SAP Community - Service Ticket Intelligence

SAP AI Business Services Tutorial Navigator Overview

2021-02-15T12:35:57+01:00

If you are familiar with SAP AI Business Services, you probably already know that they are re-usable services on the SAP Business Technology Platform providing strategic services that automate and optimize corporate processes and enrich the experience of customers across the intelligent suite.

However, did you also know that there is a range of tutorials and other useful tools that provide great guidance when getting started with AI Business Services? Let me give you a short overview of all that’s available for you!

First things first: in the Tutorial Navigator, you will find more than just tutorials. There are also Groups and Missions available.

1. Tutorials

A tutorial will provide you with a basic logically complete activity. It will take between 5 and 20 minutes to complete. Each tutorial clearly sets out the respective learning objectives and it will have a step-by-step guide with pictures. This makes it very easy to follow the whole process and it will get you started on your own project in no time! The range of available tutorials also gets extended on a regular basis, so it really is worth checking the Tutorial Navigator for updates regularly!

Here’s the list of tutorials that are available:

Data Attribute Recommendation

Use Machine Learning to Classify Data Records

Classify Data Records with the SDK for Data Attribute Recommendation

Document Information Extraction

Use Machine Learning to Extract Information from Busines Documents and Enrich Data

Use Machine Learning to Process Business Documents

Service Ticket Intelligence

Use Machine Learning to Process Service Requests

Document Classification

Use Machine Learning to Classify Documents (Trial Account)

Use Machine Learning to Classify Documents (Enterprise Account)

Business Entity Recognition

Use Pretrained Machine Learning Models to Process Unstructured Text

Use Custom Machine Learning Models to Process Unstructured Text

Invoice Object Recommendation

Use Machine Learning to Get Recommendations of Invoice Financial Objects

2. Groups

In Groups, a set of tutorials is combined into a tutorial group outlining how to build a simple application or feature. Generally, 3 to 5 tutorials on a certain service are grouped together. Tutorials are also ordered in such a way as to guide you through the whole process from start to finish. Thus, completing a tutorial group generally takes around an hour. So, Groups are great to do if you’ve got a little more time on your hands.

Here’s the list of groups that are available:

Data Attribute Recommendation

Classify Data Records with the SDK for Data Attribute Recommendation

Business Entity Recognition

Use Custom Machine Learning Models to Process Unstructured Text

Document Classification

Use Machine Learning to Classify Documents (Enterprise Account)

Use Machine Learning to Classify Documents (Trial Account)

Invoice Object Recommandation

Use Machine Learning to Get Recommendations of Invoice Financial Objects

Business Entity Recognition

Use Pre-trained Machine Learning Models to Process Unstructured Text

3. Missions

Missions are gamified activities and they combine sets of tutorials or tutorial groups to provide the basics for learning a skill. For example, the mission ‘ Use Machine Learning to Process Service Requests’ goes through the process of how to classify customer service requests and get solution recommendations – either with Postman or Jupyter Notebooks – using machine learning and Service Ticket Intelligence. As you progress through the mission you can earn badges awarded for your progress, so this makes missions fun and educational at the same time!

Here’s the list of missions that are available:

Data Attribute Recommendation

Use Machine Learning to Classify Data Records

Document Information Extraction

Use Machine Learning to Extract Information from Business Documents and Enrich Data

Use Machine Learning to Process Business Documents

Service Ticket Intelligence

Use Machine Learning to Process Service Requests

So, there you have it: a quick overview of all that the Tutorial Navigator offers. And no matter if you are just getting started with AI Business Services, or if you just want to refresh your knowledge, the Tutorial Navigator is a great place to look for input!

People sometimes wonder what the exact difference between the Tutorial Navigator and The SAP Discovery Center is, as the latter also offers Missions. I have been one of those people and if you are too, stay tuned for my next blogpost, where I explain all there is to know about the SAP Discovery Center!

PS: if you want to make sure not to miss the next blogpost, make sure you follow #Artificial Intelligence and #SAP AI Business Services and you can also follow my account on the SAP Community page!

Visit the links below to access the Tutorial Navigator and to find related topics:

Tutorial Navigator: https://developers.sap.com/tutorial-navigator.html?tag=products:technology-platform/sap-ai-business-services

For more information on SAP AI Business Services:

Explore: SAP Community Page

Dive deeper: Open SAP Course

Get an overview: Blogpost part I | Blogpost part II

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

Testing newly trained Machine Learning models on Production with Service Ticket Intelligence

2021-02-19T17:51:36+01:00

Testing a new Machine Learning model before it replaces the old one for productive usage is key to ensuring model quality and business continuity. How can we achieve this continuous testing and deployment without interrupting the existing Machine Learning model functions and operations for your business?

In the context of Service Ticket Intelligence, the service only allows for one active model per language per scenario and there's no way that you can test out newly trained models without having to de-activate the current active production model.

With the 2001 release, we introduced a test model support where you will now be able to execute tests against any trained models that are in READY (model successfully trained) state simply by specifying a model_id of interest in the inference request body payload.

Given the scenario below whereby you already have 3 Machine Learning models in your account:

Model A, id xxxA - activated and used in production via ticket application

Model B, id xxxB - newly trained model and to be tested/evaluated by an administrator

Model C, id xxxC - model not trained and not testable

To test out the new Machine learning model (model B, id xxxB) here's how you should structure your inference request body payload to make use of this feature. You will only need to provide an additional "model_id" field, referencing to the model of interest.

{

  "model_id": "xxxB",

  "business_object": "ticket",

  "messages": [

    {

      "id": 2001,

      "contents": [

        {

          "field": "text",

          "value": "you customer service is so good"

        }

      ]

    }

  ]

}

Test model support is applicable to both classification and recommendation models of Service Ticket Intelligence.

Service Ticket Intelligence is also available on SAP Cloud Platform as a free trial. Find out how you can set up your own trial account with Service Ticket Intelligence here and give the test model support feature a try today!

[Podcast] Getting down to Business with SAP AI Business Services

2021-03-01T15:49:56+01:00

Hi everyone,

I am very happy to share with you today the start of our new podcast series “Getting down to Business with SAP AI Business Services”. I am the host of this series and I am having my colleagues with me as my guest to talk about the different SAP AI Business Services. Plus, we will also have some additional, interesting topics related to the services.

In the podcast series we will be talking about the chances and opportunities organizations get when automating business processes by introducing artificial intelligence. This allows organizations to handle business operations more efficiently, have more time for value adding tasks and reduce the potential for human error. The SAP AI Business Services ensure that artificial intelligence can be easily consumed across your entire business, no matter whether you want to automate the handling and processing of invoices, e-mails, customer support tickets and or anything else.

The podcast series will introduce you to the SAP AI Business Services solution portfolio as well as several use cases for the different services. It will also give you an idea of how to get started and implement the services in your business context. We will also cover topics like the general approach to AI at SAP, how organizations using the services can gain a competitive advantage or why ethics in AI are important and how SAP is ensuring ethical AI.

Join us on our biweekly podcast to learn all about SAP AI Business and find out how automating business processes with SAP AI Business Services ease your life when it comes to automating business processes with Artificial Intelligence. It doesn’t need to be a complicated project!

Subscribe to the podcast series now on your favorite podcast platform:

openSAP Podcasts: https://podcast.opensap.info/sap-ai-business-services/

Apple Podcasts: https://podcasts.apple.com/de/podcast/sap-ai-business-services/id1553013360

Spotify: https://open.spotify.com/show/6s9WIyqajX9Ax2VW2vBdtx

TuneIn: https://tunein.com/podcasts/Technology-Podcasts/SAP-AI-Business-Services-p1407660/

Google Podcasts: https://podcasts.google.com/feed/aHR0cHM6Ly9wb2RjYXN0Lm9wZW5zYXAuaW5mby9zYXAtYWktYnVzaW5lc3Mtc2VydmljZXMvZmVlZC9tcDMv

Let us know your opinion on this podcast on your favorite platform directly or here under this blog.

For more information on SAP AI Business Services:

Explore: SAP Community Page

Dive deeper: Open SAP Course

Get an overview: Blogpost part I | Blogpost part II

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

SAP AI Business Services in the SAP Discovery Center

2021-03-05T11:31:22+01:00

The SAP Discovery Center – a (virtual) place to be. If you want to learn how to adopt the SAP Business Technology Platform and turn data into business value – this is the place!

In case you came here from my last blogpost about the SAP Tutorial Navigator and you’re unsure about the difference between the Discovery Center and the Tutorial Navigator, let’s get this out of the way first:

SAP Tutorial Navigator vs. SAP Discovery Center

First, the similarities: both, the Tutorial and the Discovery Center offer a self-learning environment for your enablement in SAP solutions for the intelligent enterprise. You can find SAP AI Business Services on both platforms.

The Tutorial Navigator offers three things: single tutorials, groups of tutorials outlining how to build a simple application or feature, and gamified activities combining sets of tutorials or tutorial groups to provide the basics for learning a skill, called missions.

The Discovery Center offers a broader cosmos of learning opportunities, use cases, expert support and customer references that quickly enable you to adopt the knowledge to launch and drive you own projects!

More about the Discovery Center

In the Discovery Center, you can choose between Mission Catalog and Service Catalog.

Mission Catalog

In the Mission Catalog you can find Missions that add business and functional value to your journey with SAP. This includes best practices and step-by-step guidance for the implementation of business use cases. When selecting the mission for a specific service, you’ll find an overview of the use case, a project board with more details, links to additional resources and links to related missions (if applicable).

For SAP AI Business Services, you can access the support section to get in touch with the product experts of our services.

In summary: Missions are all about your project, guiding you and your project team throughout the adoption phases of the SAP Business Technology Platform with free support from mission experts and a well-established support from the community!

For SAP AI Business Services, the following missions are available:

Service Ticket Intelligence: Automate Ticket Handling with Service Ticket Intelligence

Data Attribute Recommendation: Enrich your master and transactional data with AI-based data attribute recommendations
Document Information Extraction: Automate data extraction from business documents with AI and Intelligent RPA

Service Catalog

The Service Catalog offers an overview of all available services on the SAP Business Technology Platform. You can also try out the services that have mission for free. I mentioned the ones for SAP AI Business Services above.

For SAP AI Business Services, the following services are accessible via the Discovery Center:

Business Entity Recognition: https://discovery-center.cloud.sap/serviceCatalog/business-entity-recognition

Data Attribute Recommendation: https://discovery-center.cloud.sap/serviceCatalog/data-attribute-recommendation

Document Classification: https://discovery-center.cloud.sap/serviceCatalog/document-classification

Document Information Extraction: https://discovery-center.cloud.sap/serviceCatalog/document-information-extraction

Invoice Object Recommendation: https://discovery-center.cloud.sap/serviceCatalog/invoice-object-recommendation

Service Ticket Intelligence: https://discovery-center.cloud.sap/serviceCatalog/service-ticket-intelligence

The Service Catalog includes a product detail page for each of the services, which is comprised of three sections:

Features is the overview page containing lots of additional information and resources such as links to the help portal product page, customer refences, links to related learning opportunities.

Service Plan offers information on pricing and on how to order the services via the CPEA or in the SAP Store.

Related Missions will lead you to the free missions that I linked above in the section about the mission catalog.

As I said, there’s a lot to discover in the discovery center – pun intended! So, head on over and discover the cosmos of SAP AI Business Services today!

Visit the links below to access the SAP Discovery Center and to find related topics:

Discovery Center: https://discovery-center.cloud.sap/

For more information on SAP AI Business Services:

Get Insights on Historical Tickets with Service Ticket Intelligence

2021-03-12T07:28:28+01:00

Service Ticketing Systems are responsible for handling large ticket volumes to ensure a smooth and uninterrupted service delivery towards achieving business success. These tickets generally contain valuable information that one has to manually extract, in order to gain insights for further improvements.

The process is often labour intensive and costly but with the clustering feature in Service Ticket Intelligence that we released in 2201b, meaningful ticket clusters containing concise keywords and information can be automatically generated and retrieved with unstructured textual data. This will enable you to discover insights on hidden information, patterns, trends, hotspots, etc. and use it for improving the reliability, processes, customer satisfaction, etc. of your business.

The options to make use of the clustering feature are in fact limitless, but let's take a look at the common usage scenarios and find out the value that the clustering feature can bring about below.

Ticket Clustering Overview

Ticket Clustering Overall Process

Data Upload
- Historical data (tickets) is uploaded to Service Ticket Intelligence
- The data is required for training a clustering machine learning model
- A new model will be created and it will be in NEW state

Model Training
- Once data is uploaded, initiate the model training process
- Model state will transition from NEW->PENDING_TRAINING->IN_TRAINING->READY
- This process might take several minutes depending on your dataset size

Model Activation
- Once model is trained and in READY state, model activation is required before retrieval of cluster information is possible
- The model would be in ACTIVE state as soon as the model is activated

Inference
- Retrieve clusters generated
- Options are available for filtering clusters
  - top k clusters
  - group-by

Analysis and Decision
- Process the cluster information, retrieve insights with keywords, sample tickets, ticket characteristics, etc. and make follow-up actions and decisions accordingly

Examples

1. Data Upload (POST /sti/training/model)

Mapping input determines which data is to used for the generation of clusters

Mapping output determines which data is to be present in sample tickets as well as to be used for filtering clusters

Sample Request Body

{

    "scenario":

  {

      "desc":"Training Clustering travel data (small)",

      "type":"clustering",

      "language":"en",

      "business_object":"ticket"

  },

  "mapping":

  {

        "input": ["description"],

        "output": ["id", "description", "category"]

  },

  "training":

  {

        "file": "<<base64 encoded csv file>>"

  }

}

2. Model Training (POST /sti/training/model/train)

Sample Request Body

{   

    "model_id": "<<model_id>>"

}

3. Model Activation (PUT /sti/training/model/activate)

Sample Request Body

{   

    "model_id": "<<model_id>>"

}

4. Inference (POST /sti/text/cluster)

To retrieve all clusters generated from the training data

Sample Request Body

{   }

To retrieve top k clusters from the training data

Sample Request Body

{

   "options":{

      "top_k_clusters":2

   }

}

To retrieve clusters group-by field from the training data

Sample Request Body

{

   "options":{

      "cluster_groupby":{

         "column":"category",

         "value":[

            "compliment"

         ]

      }

   }

}

Additional Resources

Postman collection: https://github.com/SAP-samples/service-ticket-intelligence-postman-collection/tree/clustering

API specs for clustering training: https://help.sap.com/viewer/5088c3bb02144e7782959bb1529ca70e/SHIP/en-US/d663408e81b6406f98bb2b14b45f3e6c.html

Examples: https://help.sap.com/viewer/5088c3bb02144e7782959bb1529ca70e/SHIP/en-US/d2b81fc4b8424a0eaa9d6cf420751fe9.html

API specs for clustering inference
https://help.sap.com/viewer/5088c3bb02144e7782959bb1529ca70e/SHIP/en-US/5d7142a96c0f49428323fa75a0a40a3a.html

Examples
https://help.sap.com/viewer/5088c3bb02144e7782959bb1529ca70e/SHIP/en-US/61fdc051905149e398c23d33a9cb8824.html

To try out other features (classification, recommendation scenarios) of Service Ticket Intelligence, we are also available on SAP Cloud Platform as a free trial.

Find out how you can set up your own trial account with Service Ticket Intelligence here and give the service a try today!

For more information on SAP AI Business Services:

Explore: SAP Community Page

Dive deeper: Open SAP Course

Get an overview: Blogpost part I  | Blogpost part II

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

[SAP AI Business Services Success Stories] How an intelligent Service Cloud helps B&D improve the efficiency of customer support processes

2021-04-15T12:38:41+02:00

About B&D

B&D markets and manufactures garage doors and electronic openers for residential, commercial, and industrial use.

B&D is part of DuluxGroup Ltd., one of the most trusted brands in Australia and New Zealand, which manufactures, markets, sells, and distributes premium branded paints and coatings that protect, maintain, and enhance the places in which we live and work.

The challenge

DuluxGroup Ltd. wanted to streamline customer support processes by implementing new, next-generation technologies to onboard its contact center representatives quickly and easily. The high expectations of their customers had to be met in addressing and resolving customer support needs.

SAP AI Business Services Solution

Service Ticket Intelligence - using the embedded ticket intelligence capabilities in the SAP AI Business Services portfolio with the SAP Service Cloud solution, enterprise edition.

Service Ticket Intelligence is part of the SAP AI Business Services portfolio. Leveraging deep learning neural networks trained on large amounts of historical data, Service Ticket Intelligence helps build a self-driven customer service powered by machine learning. The service classifies incoming service requests automatically, as a basis for routing them to the right agent. The model understands the semantics of unstructured ticket messages and improves over time with more processed service tickets and user feedback.

Side note: In spring of 2021, a new beta feature for Service Ticket Intelligence was released. Now, meaningful ticket clusters containing concise keywords and information can be automatically generated and retrieved from unstructured textual data. To find our more, read this blog post by my colleague Samuel Seah or visit the SAP Help Portal.

Achievements

DuluxGroup Ltd. automated the categorization of incoming tickets. Service agents are now provided with recommended solutions to improve operational efficiency, allowing for much quicker problem solving and service response time. The text classification feature predicts categories of incoming tickets based on historical service tickets that have been classified previously so that customer service agents can address and resolve customer issues quickly.

With the support of SAP AI Business Services, DuluxGroup’s B&D Group achieved significant automation improvements:

70% automatic ticket classification

5x improvement in first response times

85% model accuracy using 150,000 historical records across 16 categories

Faster onboarding for contact center representatives

At B&D, less time is now spent on manual tasks such as sorting e-mails. The value creation that was achieved through automation freed support representatives to focus on resolving customer issues quickly.

Visit the links below for more information

Customer Reference: Link

For more information on SAP AI Business Services:

Explore: SAP Community Page

Dive deeper: Open SAP Course

Get an overview: Blogpost part I  | Blogpost part II 

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

[SAP AI Business Services Success Stories] One service, two customers: How Service Ticket Intelligence enables the intelligent enterprise

2021-04-30T11:35:16+02:00

Internal ticket handling can be an issue for many companies. Often, ERP-System are complex and thus, IT ticket response time is slow due to the high manual effort involved. In turn, this keeps users waiting to get their issues resolved. Additionally, many companies still operate on costly and ineffective work processes. Even if businesses want to speed things up, the increased pressure to do more with less (less people, less time, less operational budget) can stand in the way of making the necessary progress.

Of course, not an ideal starting situation, but one that can be fixed by using combined intelligent services.

Two companies that have done exactly that are NHK Spring Co, LTD and Hewlett Packard Enterprise (HPE).

Both firms use a solution from the SAP AI Business Services portfolio: Service Ticket Intelligence, a microservice on top of SAP Business Technology Platform with flexibility to integrate with various SAP and non-SAP offerings.

Service Ticket Intelligence is part of the SAP AI Business Services portfolio. Leveraging machine learning algorithms built using historical service interaction, Service Ticket Intelligence improves the efficiency for ticket handling of external customer service centers and internal shared service centers. It supports the service team's ability to handle larger volumes of service interactions by providing automatic ticket classification and recommended solutions.

For fully integrated scenarios and to increase the automation rate of business processes, the service can also be combined with SAP Intelligent Robotic Process Automation and SAP Conversational AI.

If you are interested in more details about Service Ticket Intelligence, you should check out these two blogposts by my colleagues:

An Overview of Service Ticket Intelligence

What the AI future brings: Highlights from the first SAP AI Business Services Focus Week

NHK Spring Co, LTD.

offers quality spring suspension solutions, operating a global group network out of Japan. NHK implemented Service Ticket Intelligence to reduce manual ticket management and communication. Now their chatbot and ticketing system provides automatic incident user proposals based on the information given initially. Assisted by SAP Innovation Services, the implementation took only 2 months. With the help of Service Ticket Intelligence, NHK Spring Co, LTD. significantly improved internal ticket handling efficiency and subsequently user satisfaction.

Hewlett Packard Enterprise (HPE)

is a global, edge-to-cloud and platform-as-a-service company. HPE used the business operations self-healing service as part of SAP MaxAttention™ services to simplify its IT processes. Automation was supported with a prepackaged AI platform enabled by SAP® Intelligent Robotic Process Automation, SAP Conversational AI, and Service Ticket Intelligence. The off-site delivery was 100% successful although it happened during COVID-19 pandemic! Now, a system that restarts completely autonomously, using chatbots and robotic process automation, has been implemented. Tickets are categorized and routed to the right department with the help of a chatbot integrated with the third-party Service Now platform. As a result of the improvement in internal ticket handling, HPE can now provide a smooth customer experience.

There you have it, two very interesting stories showcasing the versatility of Service Ticket Intelligence!

By the way:

In spring of 2021, a new beta feature for Service Ticket Intelligence was released. Now, meaningful ticket clusters containing concise keywords and information can be automatically generated and retrieved from unstructured textual data. To find out more, read this blog post by my colleague Samuel Seah or visit the SAP Help Portal.

Visit the links below for more information

Customer Reference: Link

For more information on SAP AI Business Services:

Explore: SAP Community Page

Dive deeper: Open SAP Course

Get an overview: Blogpost part I | Blogpost part II

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

[SAP Community Call] Ticket Clustering of Service Ticket Intelligence

2021-06-01T10:44:59+02:00

Dear SAP Community,

The SAP AI Business Services team is happy to announce another SAP Community Call!

Overview

Title: Ticket Clustering of Service Ticket Intelligence

Date: June 15, 2021

Time: 11:00 am CET / 03:30 pm IST / 06:00 pm CST / 07:00 pm JST

Duration: 40min (including Q&A)

Summary:

We will be introducing the new feature of Service Ticket Intelligence - Ticket Clustering - in this session. You will first get to know the feature in detail, its data science details, the potential use cases with value, followed by a demo. Additional links will be provided if you are interested to know more details.

Agenda

Short introduction of Service Ticket Intelligence

Ticket Clustering: the feature and data science details

Use cases

Demo

Q & A

Speaker

haodan.yang#overview, Product Manager SAP Artificial Intelligence

Register now!

You can simply register from here or you can go to the registration page for our SAP Community Call directly. In case you cannot attend, you can watch the recording on the SAP Community Calls home page afterwards. Of course, I will also be updating this blog post after the call to include the session recording.

Looking forward to seeing you on the call!

Do you want to start right away? No problem:

Explore: SAP Community Page

Dive deeper: Open SAP Course

Get an overview: Blogpost part I | Blogpost part II

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

How hyperautomation can optimize financial shared service centers using SAP AI Business Services

2021-08-04T13:15:08+02:00

Introduction

Shared service models together with process standardization and centralization have changed the efficiency and effectiveness of modern financial service organizations. In today’s world, CFOs are heavily investing in automation technologies such as Robotic Process Automation (RPA) and Artificial Intelligence as a next step for enterprise transformation.

Finance processes such as Accounts Payable, Accounts Receivable, and Expense Management can often be further automated, thus freeing up employees to focus on value-adding activities. Gartner predicts that two-thirds of all customer service interactions will no longer require a human intermediary.

A few weeks ago, I presented a guided demo on “Automating financial shared service centers with SAP S/4HANA, SAP Intelligent Robotic Process Automation and SAP AI Business Services” as part of the Hyperautomation Webinar Series.

Use Case – Vendor Inquiry Process

In this webinar we looked at one specific finance process – the vendor inquiry process. SAP customers with shared service organizations often handle a high volume of inquiries from vendors.

Request Handling in Shared Service Centers

As a first step, service tickets are created automatically from incoming email requests and enter the ticket inbox of the shared service center agent. Next, Service Ticket Intelligence processes and classifies these service requests.

Business Entity Recognition identifies and extracts predefined entities from the service request itself and Document Information Extraction extracts structured information from attachments. The information extracted from Business Entity Recognition and Document Information Extraction (for example, the invoice number) is then checked against the relevant invoice posting status in SAP S/4HANA.

This status is the basis for either providing the service agents with assistance (e.g. preparing a response), or even directly answering the inquiry via a SAP Intelligent RPA bot automatically.

SAP AI Business Services involved

Three distinct SAP AI Business Services are involved in this scenario.

Service Ticket Intelligence classifies and processes incoming service requests. It leverages deep learning neural networks trained on large amounts of historical data. The model understands the semantics of unstructured ticket messages, classifies the ticket into their most likely categories and recommends to the agent solutions or knowledge-based articles from similar previously answered tickets.

Document Information Extraction extracts structured information from unstructured documents. The extracted information can be automatically enriched with your existing structured master data and transactional data.

Business Entity Recognition helps you to detect and highlight any given type of named entity in unstructured text and classify it in accordance with predefined categories.

SAP AI Business Services for shared financial service center hyperautomation

By automating and optimizing the request handling process in shared service centers, with the help of SAP AI Business Services, SAP Intelligent RPA and SAP S/4HANA, businesses and organizations can achieve new levels of operational excellence.

You can watch the full-length video of my session here:

Further information on SAP AI Business Services:

Explore: SAP Community Page for SAP AI Business Services

Dive deeper: openSAP Course & openSAP Podcast

Get an overview: Blogpost part I | Blogpost part II

Exchange Knowledge:

Document Classification Questions | Document Information Extraction Questions

Business Entity Recognition Questions | Service Ticket Intelligence Questions

Data Attribute Recommendation Questions | Invoice Object Recommendation Questions

How to Implement Intelligent AI Use Cases for Your Business

2021-09-20T16:04:05+02:00

Hello SAP Community,

Either you're new or advanced in the artificial intelligence topic, our new use case repository is a great and simple way to learn how AI solutions from SAP can help you build an intelligent enterprise using prebuilt AI and data-driven cloud applications.

Discover how organizations are using AI to solve complex business challenges in:

Finance

Sales & Commerce

Procurement

Customer Service

The AI portfolio from SAP includes SAP AI Business Services and SAP Conversational AI, both available on SAP Business Technology Platform (SAP BTP).

See concrete examples of how to consume these products natively as part of SAP standard applications (e.g SAP S/4HANA, SAP SuccessFactors...) or extend your own SAP systems.

Here's a sneak peek:

You can deep dive into each use case to find more information about:

The business context

Challenges faced by organizations in this context

The solution provided by SAP

Main benefits

Further learning resources like demos or product pages

Ready?

Explore our AI use case repository now!

We'll add additional use cases in the next following weeks, so stay tuned.

Eager to learn more?

Join our AI community

Request a demo

Automate Your Customer Service By Infusing Artificial Intelligence Into Business Processes

2021-10-01T11:15:09+02:00

At SAP, our mission is to build intelligent enterprises by infusing our AI technologies into applications and business scenarios across all lines of business (HR, procurement, or finance), at scale.

In this blog post, let's focus on two concrete AI applications in the customer service area.

Use Case #1 - Chat Optimization for Ticket Handling

As customers and employees increasingly communicate using chat technologies, users expect these tools to deliver intelligent responses. They also want companies to resolve their queries right away, within the chat window.
To help create a rewarding user experience, organizations need to automate chat activities.

What are the challenges faced by organizations today?

1/ Rising number of customer and employee conversations taking place in chat windows

2/ High user expectations for intelligent responses in the chat, including a simple way to resolve queries

3/ Need for a more engaging user experience during chats

What is the solution offered by SAP?

1/ An AI-powered chatbot that directly answers customers without requiring effort from agents

2/ Service ticket intelligence service that automates ticket classification, proposing solutions to facilitate the ticket-handling process

What are the benefits for our customers?

1/ Enhanced customer experience using solution proposals from past requests

2/ Smoother customer interaction with chat

3/ Reduced number of service tickets

4/ Streamlined customer service operations

Learn more about the AI-powered services used in this example, all available on SAP Business Technology Platform:

SAP Conversational AI chatbot building platform

Service Ticket Intelligence

Dig deeper by watching this success story with our customer NHK Spring, one of the world's leading spring manufacturers:

Use Case #2 - Service Request Handling

Thanks to digital transformation, customer interaction centers are responding to ever-higher volumes of requests. In shared service centers, agents must handle these requests across multiple channels, delivering more responsive service despite growing workloads.
To improve service levels, companies must automate rote processes, freeing knowledge workers to handle more difficult issues.

What are the challenges faced by organizations today?

1/ An increasing speed and volumes of customer inquiries processed by shared service centers

2/ Multiple communication channels that must be managed by agents

3/ High workloads and queued worklists

4/ A stressful work environment for agents trying to provide responsive service

What is the solution offered by SAP?

1/ An intelligent AI service that automates the classification and processing of customer service requests (see Figure 10)

2/ Machine learning models that categorize customer input from a variety of channels into internal service groups

3/ Automatic identification and extraction of business entities from the content of requests

4/ Automatic processing of standard inquiries supported by intelligent RPA technology

What are the benefits for our customers?

1/ Faster, more accurate responses to customers

2/ Fewer repetitive tasks for service center agents

3/ Greater focus on valuable creative tasks

Learn more about the intelligent technologies used in this example, all available on SAP Business Technology Platform:

Business Entity Recognition

Service Ticket Intelligence

SAP Intelligent RPA

Check this 4 minutes demo to see it in action:

These are just two successful examples of how artificial intelligence can significantly improve CX at scale when integrated into business processes.

Visit our AI use case repository to see other examples 🚀

We’ll add additional use cases in the next following weeks, so stay tuned!

Ready to implement AI in your business?

Contact our team

Join our AI community

Building a Dashboard for ML based Ticket Categorization analysis

2022-02-01T23:49:19+01:00

Using Machine learning for Service Ticket categorization

Machine learning for Service Ticket categorization is a powerful tool to help automate ticket categorizations. It is based on the prediction capabilities of Leonardo which learns from your data and uses that to predict the ticket category and priority with a certain degree of confidence.

You could have a catalog using Service Categories, Incident Categories, Object Categories and Resolution Categories. Depending upon how you have set up your catalog – in a flat structure or hierarchical structure, you could specify at what level(s) ML predictions should happen. This takes place with the Ticket Categorization model that you can train and activate. You can specify whether you like Priority to be predicted as well. Apart from this setting, you can also set a threshold. If ML’s prediction confidence is above the threshold set by you, it can automatically populate the ticket category and Priority fields - thereby freeing up resources and saving time of the service agent who manually classifies the incoming tickets. For example, if you’ve specified the threshold as 60, then only if the confidence is greater than 60, the system auto-populates the relevant category field in the ticket

KPIs that matter for analyzing ML effectiveness

As an organization, if you have implemented or are in the journey to implement Service Ticket categorizations using Machine Learning in Service Cloud, I’m sure you’re also wondering how to analyze its effectiveness across different regions or support teams

Some of the KPIs that you probably want to track are

Volume of incoming Tickets

Number of tickets that are getting predicted above the threshold

Incoming tickets vs ML predicted tickets per team / country / region etc

Top X predicted categories

Prediction trends

True value vs Predicted value

Not having visibility into these aspects pose the challenge of not being able to understand the effectiveness of ML based ticket categorization for your organization, explaining the value internally to the management or even taking decisions to clean up your catalog or to retrain the model

Aim of this blog

This blog aims to inspire you with ideas to create your own dashboard for the ML based Ticket Categorization scenario, using standard Business Analytics capabilities. I will introduce key aspects of the Data sources involved, key figures to create, report views and finally the dashboard

Design and setup

As always, the best practice recommendation is to first think and decide on paper about what you want to achieve with the dashboard. What tiles make sense and how it will help you in the analysis. Then work backwards to see what you need to achieve that - which data sources, key figures and reports are required. Which ones are available in standard, and which ones do you need to create as custom

Example setup from my tenant

Prediction enabled for Incident category, Process category and Priority

Threshold set at 60

What do I want in the dashboard?

For my dashboard design I want two broad perspectives

Overall Ticket volume and prediction information – by team and date/month

Deeper dive into Incident category predictions

To achieve this, I went back to the KPIs that were important as explained earlier. I looked into what the Ticket Intelligence data source offers and what it doesn’t. While the DS has the main aspects – information of all actual categories as well as ML predicted categories, ticket information etc. as characteristics, it doesn’t have the Created-on date, or the service and support team information. The DS also has a counter and prediction confidence as key figures. What I needed additionally was a way to get a count of all tickets with prediction confidence greater than the threshold and reflect it as a % of the total volume of tickets. Ticket threshold is not available in the data source, but this is not something one would change frequently. Hence, I could still use it as an absolute value in my calculations

Data Source:

For reporting ML predictions for Ticket Categorization – Ticket Intelligence data source

Purpose of this data source – It captures the predicted values as well as the confidence value of each ticket, provided the confidence is greater than the threshold. Hence, say you set the threshold as 60, but the prediction confidence for Incident Category was 55, then it wouldn’t get reflected in the data source. However, if you had selected Incident category as well as Priority for prediction, and the prediction confidence was 55 for Incident category but 85 for Priority, both would get reflected in the data source. This is since at least one of them was greater than 60

Data Source Fields:

Do a preview of the data source to see the fields. The data source gives us information at a ticket level. This includes the actual as well as the predicted values on that ticket. If populated (as explained above), it tells you the confidence

Key Figure design

For the KPIs explained above, I ended up creating the following key figures using the standard key figures in the formula

Predicted Incident Category count Greater than Threshold

Predicted Service Category count Greater than Threshold

Predicted Priority count Greater than Threshold

% of tickets with Incident categories predicted above the threshold

% of tickets with Process/Service categories predicted above the threshold

% of tickets with Priority predicted above the threshold

Here’s an example:

Key Figure: Predicted Incident Category count Greater than Threshold

Step 1

Step 2

Step 3

Explanation -

The boolean condition in step 2 works on each ticket to check if the prediction confidence was greater than equal to 60. If yes, it would return True (1) else False (0)

Step 3 collects all the tickets via exception aggregation and applies a Total on that. For example, if there was a total of 1000 tickets out of which 700 had an incident category prediction confidence GE 60, it would sum up the 1s(True) for these 700 tickets and 0s(False) for 300 tickets. Thus, you would get an aggregate/total of 700. This is required for the % calculation key figure shown below

Key Figure: % of tickets with Incident categories predicted above the threshold

Step 1

Step 2

Report design

Finally, I created a new report on the Ticket Intelligence data source and pulled in these key figures. For the missing characteristics, I preferred to pull them in using Add fields instead of creating a Join with the Ticket data source. I used these additional fields as selection parameters - especially the created on field. Having a selection and making it mandatory is recommended because with time the data volume can be huge and you probably want to restrict that for more near term analysis

Also, note that this dashboard is built to show aggregated information. If you have implemented ML you probably have large volumes of data. I built the report on the Ticket Intelligence data source without any joins. The selections for the date were on the Created On field which was pulled in using Add Fields.

Standard and Custom Key figures available in the report

Report Views: I then created the views I needed for my dashboard. You will see these as part of the dashboard tiles

Here’s a Tip! Follow a naming convention that allows you to easily locate these views in the dashboard design process. You can always rename it there. I followed a prefix 010 (don’t ask me why!)

Dashboard design

Finally, it was time to design the dashboard and to present the perspectives visually

Scroll up to see the section What do I want in the dashboard?

Knowing that I can have a maximum of 3 * 3 tiles in the dashboard, I decided on the layout – first two columns for perspective 1 and the last column for perspective 2. This is the dashboard that I had on paper and that I finally managed to achieve

Perspective 1: (Reads top to bottom)

Shows the Ticket count per team – This is important since you can see the incoming volume of tickets per team. You could also use regions, countries, service orgs etc

Prediction efficiency chart – Shows the % of automation of tickets. Wherever the prediction confidence was greater than 60 (that was my threshold), it represents it as a % of the total ticket count

Overall prediction counts – Organizations also want to see the counts, not just the % to get a true idea – this stacked chart shows the incoming ticket count vs predicted incident category count, predicted service category count and predicted priority count

Finally, the line chart shows the prediction efficiency per month. This could especially be helpful in the initial months after the rollout to see if the graph is on an upward or constant trend. Downward is not desirable

Perspective 2:

Focus on Incident category prediction counts per team – this is helpful to gauge how useful is the ticket prediction for individual teams

Top 5 categories predicted – useful again to see which categories are getting predicted the most. If bulk of your issues are in these top 5/top x categories, you’re good

Finally, there is an incident category comparison matrix – this is a report view which presents actual/final Incident categories maintained on a ticket vs what was predicted by the system. It’s a true label vs predicted label comparison in that sense. Here’s a closer look into it

We can see that for the Actual label Missed Appointment there are 109 ML predicted Missed Appointments as well. But there are 2 predicted as Fridge app OS version and 7 which were not predicted by ML. If a large portion is correct – meaning where the actual vs predicted value is the same, then the quality of prediction is also good. If it’s the reverse, then there’s a need to analyze what the data set, volume and data balance looks like. Maybe there is also a need to retrain the model or clean up the catalog structure

You can also export the matrix data into excel (ideally without the columnar layout) and build further calculations to report on the prediction quality. Here’s a simple example where the data was exported and a new column D was introduced to get a total count of true labels vs false labels in column E. Finally, column G shows a % of prediction accuracy based on this calculation. You can build good visualizations of course using standard excel features. You can also use the Excel Add in for this – it’ll allow you to have real time data from Service Cloud and you can also access the report directly from within Service Cloud

This dashboard is also interactive in nature and offers global value selections (Created on, Support team). Interactivity allows you to focus your analysis - I built this interactivity on the Support team. Selecting a specific team or teams makes the other tiles react - it basically passes the Support team as a filter to the other tiles

To summarize, you can see how I used the Ticket Intelligence data source to create my own key figures, report and dashboard for analysis of the predictions.

Hope this gives you some ideas about your own Dashboard implementation for ML based Service Categorization

Here are a few helpful links that may be relevant for your Sales/Service Analytics and ML journeys

Machine Learning in Sales and Service Cloud Playbook

Deliver Service Excellence with the power of AI Webinar

Drive Sales by smart decision making with Analytics Webinar

Service Analytics for smart decision making Webinar

Adoption Lab – Sales Analytics Webinar/Adoption Lab

Cheers !

Vinita

SAP Intelligent Adoption & Consumption Team

Optimize Ticket Handling with Service Ticket Intelligence (SAP Hyperautomation Webinar Series 2)

2022-03-01T10:37:32+01:00

Last month I had the pleasure of participating in SAP’s second Hyperautomation webinar. During my session, I showed how SAP Process Automation technology can optimize the ticket handling process for customers using SAP AI Business Services.

This scenario is particularly relevant to all of us during this COVID situation because so many people, including customer service representatives, are working from home. This means that almost all customer service centers are now online, and it has also meant a huge increase in the number of tickets received and in the amount of pressure on customer service agents to deliver exceptional customer service experiences.

But customer service agents at home are finding it hard to deliver high-quality service because they often don’t have ready access to real-time channels for getting feedback and guidance from more experienced service agents nor even to knowledge-based reference articles. This creates a big gap between the increase in customer expectations and agents’ ability to handle the overwhelming number of customer service requests in an efficient, complete, and satisfactory way.

During my session, I presented step-by-step about how to use SAP AI Business Services, which includes Service Ticket Intelligence, Business Entity Recognition, and Document Information Extraction, to help customer service agents maximize ticket handling efficiency.

What happens when a new ticket comes via email or another kind of online channel? First, Service Ticket Intelligence will propose potential categories for the ticket, identify the language, perform a sentiment analysis (that is, look for key words to determine if the ticket is a complaint, a request, or a compliment), and propose a priority prediction for the ticket.

Business Entity Recognition can identify such entities as the product ID, the date, the location — in short, all the relevant information in the ticket. And then, based on the ticket content, Service Ticket Intelligence will recommend a similar ticket that was handled in the past for the agent to use as a reference when handling the current ticket.

If the ticket comes with any attachments, such as an invoice sent as a PDF, for example, Document Information Extraction convert the attachment to plain text and then extract the invoice ID, the date, the vendor ID, and other relevant information.

All of this automatically occurs when the ticket arrives in the ticketing system so that the agent will have instant access to all this information instead of having to extract it manually. This means they can handle the ticket faster and get back to the customer faster, which will make the customer happier as well. So this is an end-to-end scenario for ticket handling that leverages AI solutions.

Further, SAP Service Ticket Intelligence also provides a ticket clustering feature that you can use to identify what kinds of problems are recurring within your system or product, and then use that information to improve customer-facing processes or features.

I hope you will watch this session and come away with some ideas about how you can use SAP AI Business Services to improve not only the experience your customers have, but also the experience that your service agents are having during these challenging times.

Below are some quick links in case you want to learn more and try out these services.

Follow SAP AI Business Services on SAP Community

Get started with Service Ticket Intelligence

Learn more about how to implement Ticket Clustering

Browse our AI use cases including Service Ticket Intelligence

Use Insights to Create Business Value with Ticket Clustering of Service Ticket Intelligence

2022-03-02T08:16:45+01:00

Computers never have a problem understanding structured datasets, but for unstructured data like free text, it is always a pain. With AI and machine learning technologies, we are now able to tap into the richness of information stored in the unstructured datasets.

For instance, service centers of big companies usually receive thousands of support tickets per day. Even though there’s a lot of valuable information captured in the ticket content that could potentially be used to know better about some customers or improve the product/service itself, it’s impossible for a human being to go through all these tickets just to generate insights with statistics.

That’s where the power of Natural Language Processing (NLP) text clustering comes in!

Envisioning the high potential, Service Ticket Intelligence (STI) – part of the SAP AI Business Services portfolio, recently had a GA release (2002a) introducing a new machine learning-based feature called Ticket Clustering. This new capability is aiming to generate insights from the content of voluminous tickets received by service centers, and generate actionable business value out of these insights.

In this blog post, I’d like to show you how we could get insights from unstructured text, by following a real business use case.

Use Case Example

Company A has a big customer X that generates over 2300 tickets received from January 2019 to February 2020. The account team of customer X would like to get an analysis of the tickets to better understand the customer feedback and challenges encountered, so they can plan what and how to pitch to them with new offerings.

These tickets get uploaded to Service Ticket Intelligence using the Ticket Clustering feature and 100+ clusters are generated with a list of keywords attached for each cluster to show their meaning. The team arrange the clusters in sequence with the descending number of tickets, and realize that the top 5 clusters account for over 54.07% of total hours for resolving all tickets, and over 53.10% of overall tickets counts.

With that, the account team focus on these 5 clusters and start to take a closer look based on the keywords provided and sample tickets that fall under each.

Here are the two main insights pointed out after the analysis:

Insight 1: Cluster 5 (contains 73 tickets with 249 hours spent to handle the tickets) is about access (role) request and password tools.

Insight 2: There are two main descriptions identified among the top 5 clusters, mostly referring to manual monitoring, which can be technical, MOCC (Managed Operations Control Center technical alert handling), and APO.

Two follow-up actions have been identified by the team in charge:

Action 1: The service center manager proposes to create a self-service/automated access (role) request and password tools, which will potentially have the following benefits:

- Up to ~73 total tickets per year can be reduced

- Up to ~249.41 total hours for ticket handling can be saved

Action 2: The account team understands customer’s challenges on manual monitoring, and comes up with a sales play around it to better tackle customer needs.

To sum up, the clusters formed out of these 2300 tickets by the Ticket Clustering feature of Service Ticket Intelligence have been used to generate concrete action plans and achieve tangible business benefits.

There are many more use cases where the Ticket Clustering feature can be leveraged in enterprise applications to create business value, and ultimately improve the customer experience.

Stay tuned with Service Ticket Intelligence and SAP AI Business Service by following these useful links below, and let us know if you have any comments in the section under the blog post!

Follow SAP AI Business Services on SAP Community

Service Ticket Intelligence Documentation via SAP Help Portal

Blog on Get started with Service Ticket Intelligence

Learn more about how to implement Ticket Clustering

Browse our AI use cases including Service Ticket Intelligence

Meet SAP's AI Portfolio and What It Can Do For You

2022-03-29T08:16:27+02:00

When I was 8 years old, my dad and I built a robot out of wood. Unfortunately, we couldn’t agree on a suitable head for the robot. My dad wanted to just put an empty wooden box as a head. I on the other hand wanted my robot to be smart. I wanted it to talk to me and play with me. Our project came to an end. But I didn’t give up. I used an empty cereal box and put my Walkman inside for some ‘interaction.’ Very unsatisfying…

Years later at university I developed my first neural network, and I was amazed! My love for AI was reignited. Another decade later I found myself at SAP where I now get to be a Developer Advocate for the Artificial Intelligence portfolio. I am so excited to share with you all the possibilities AI has to offer for your applications and your business! I hope you learn to love it as much as I do.

Without further ado let’s meet SAP’s AI portfolio:

SAP's AI Technology Portfolio

SAP AI Business Services

SAP AI Business Services are reusable services with underlying machine learning models that tackle some of the most common business problems. The models are pretrained and/or customizable, which means you can use them as they are, but you can also retrain them with your own data. You can access the services either through the RESTful APIs (Swagger UI - example) or via a Python SDK. The services are currently being migrated to SAP AI Core (SAP’s AI runtime - I will explain it further down) which is why you can also already access some services via AI API and SAP AI Launchpad. There are currently four SAP AI Business Services:

The Business Entity Recognition service (BER) can detect specific entities in your unstructured text. Within the service you will find four machine learning models, each specialized in identifying different entities. These span from finding the address, invoice header or general terms in a text. Using this service will help you to speed up the process of manually looking through your text data from, for example, emails or service tickets. BER is a customizable service, which means you can optimize it for your own text data.

Example use case business entity recognition

With the Data Attribute Recommendation service (DAR), you can classify your data records and predict missing data points. DAR is also a customizable service using business blueprints, which is why the Invoice Object Recommendation service is now available as a business blueprint in DAR. Check this blog post to find out more about business blueprints and all the benefits that come with combining the two services.

The Document Information Extraction service (DOX) uses documents as an input and extracts the text into structured data. By combining OCR (Optical Character Recognition) with post processing of the extracted text and data enrichment, DOX can significantly reduce the time needed to process incoming documents. While Business Entity Recognition looks specifically for certain entities in text, DOX extracts specific information from a file including their position as coordinates in the given document. Business Entity Recognition is already used in DOX for certain use cases.

With the Personalized Recommendation service (PR) you can give users recommendations based on their browsing history and your item descriptions. You need to train a model for each of your websites. The trained model can then provide similar-item recommendations and return smart-search results. The tentative release date of PR is currently April 18th.

Personalized Recommendation we all know from Online Shops

SAP AI Core

SAP AI Core is not only the (soon-to-be) underlying runtime of the SAP AI Business Services but also has a lot of functionalities of its own. You can use it to train your own models or deploy already trained models. It comes with the AI API which is a standardized interface to access all your models on SAP AI Core. It is a service on SAP BTP, Cloud Foundry environment. You can realize any machine learning use case on SAP AI Core you’d like by using open-source frameworks.

SAP AI Launchpad

SAP AI Launchpad is the platform to manage all the magic. The SAP AI Launchpad is a multitenant SaaS in SAP BTP (again within the Cloud Foundry environment). You can use it to manage your AI use cases across multiple instances of AI runtimes. It is your admin app!

Screenshot of SAP AI Launchpad (UI)

SAP HANA PAL

SAP HANA Predictive Analysis Library (PAL) is a collection of several machine learning algorithms that you can access in HANA, ranging from time series analysis to classification or regression algorithms. If your data is already in SAP HANA and SAP HANA PAL offers an algorithm to solve your machine learning case, then it is an easy to implement and powerful way to do machine learning at SAP!

Intelligent Scenario Lifecycle Management (ISLM)

ISLM is as the name suggests a Machine Learning lifecycle management tool for SAP S/4HANA applications. So it is basically the SAP AI Launchpad of S/4HANA. ISLM has two scenarios. The Embedded scenario (using for example HANA's PAL algorithms) or the Side-by-Side scenario by connecting it to for example an SAP AI Business Service.

SAP Conversational AI

SAP Conversational AI is our chatbot solution! For news, videos and more information on the topic follow my wonderful colleague daniel.wroblewski.

Closing

Some of you might wonder what happened to SAP Leonardo. Well as cool as some of the services were, SAP’s strength lies in our knowledge of business use cases! With our new portfolio we still offer you a way to do almost everything you could do with Leonardo but with a lot more possibilities, transparency, and flexibility. Additionally, we offer the SAP AI Business Services which are especially tailored to your business problems!

I hope I was able to clarify what’s behind AI at SAP and get you hooked on our AI portfolio. In the coming months I will take a closer look at each of the SAP AI Business Services, SAP AI Core and SAP AI Launchpad. I will show you how to use them, what’s behind the scenes and share some cool and creative use cases with you! If you want to learn more about AI at SAP subscribe to our YouTube channel and follow me here and on social media! If you have any questions, write it in the comments or reach out to me!

Helpful links:

Video on SAP's AI Portfolio

SAP Help Page - Artificial Intelligence

SAP Community - AI Business Services

Blog Post on SAP AI Business Services Tutorials

SAP Help Page - SAP AI Core

SAP Help Page – HANA PAL

Women in Artificial Intelligence at SAP #2: Get to Know Haodan Yang

2022-06-10T16:26:20+02:00

Recently, we kickstarted a new initiative to put the spotlight on women who are working in the technology industry and, more specifically, in artificial intelligence at SAP.

Through the insights jana.wuerth, we had the opportunity to learn how she handles go-to-market strategy and commercialization tasks in the AI space - a true inspiration and motivational injection for anyone who wants to discover how women embrace technology.

However, achieving actual change and progress can only happen if we are constantly open to new impulses and fresh ideas. Therefore, we will now continue with another perspective: the one from our colleague haodan.yang, who will tell us what it is like to take over a bridge role that requires deep technical and business knowledge at the same time.

As a senior product manager at SAP, she is responsible for connecting the development team and all business stakeholders for the products (including SAP internal teams who embed our AI solution, SAP sales ecosystem, and external customers) across the whole product lifecycle.

Let's start right away with an exciting question: what specific skills have helped you the most to get where you are today?

Communication skills. As a product manager, I need to talk to anyone involved in the product development lifecycle, ranging from the product development team to go-to-market experts to sales colleagues and customers who give feedback on how to improve. It's a wide range of people with diverse backgrounds, and I need to talk to them in the languages they can understand for effective communication and a successful product.

I think it is also critical that I clearly know what I enjoy doing and what I'm good at. I like to talk to people since I was a kid. I remember that I was always the one who got caught in class talking to my classmates. And along my career path, I had doors opening for me to do more research-oriented or technical jobs. Still, eventually, I followed my heart and chose to become a product manager, where talking to people is a day-to-day job. Find your passion and turn it into a career.

Did you expect to pursue a career in artificial intelligence, or did you have other plans? What initially sparked your interest in this topic?

I got to know Neural Networks and AI when I was doing my master's program and liked it. But I have never planned a job in the AI domain to start with. When I took the course back then, AI was not a popular term as it is today. But what I knew about myself was that I always enjoyed the innovation-related topics, the LEAN Startup approach, and all the possibilities of cutting-edge technologies. I did three startup projects during my master's program, and when I started job searching, SAP Research and Development center caught my eye. Not long after I joined the team, SAP started its AI journey in the team. Now that I look back, I feel it is all connected.

Was there an inspirational female person in your life that motivated you to focus? Or a particular event?

When I was doing the startup projects in school, I really admired Prof. Luda. She used to head the product development team at GE and headed the Enterprise Development Lab under the Faculty of Engineering at the university back then. I was a big fan of her class, where she could elaborate the business concepts easily to any audience and build the linkage to the high-tech products. Since then, I think it has been my dream to perform a bridging function. She's a true role model.

Would there be best practices to follow in order to help women at any stage of their professional development – especially in the tech space?

First of all, I don't think there is some best practice that fits everyone or any stage of their career. What's essential in the tech space is to keep an eye on what's happening and what's the next to come. Technologies and things change so fast, what works today might be outdated tomorrow. So, I feel it's important to keep up refreshed and synchronized with the world.

Aside from that, I feel women did have some unique advantages in the tech domain. While men are generally more hands-on in their own world, I think women have better capability to sympathize and coordinate things. What I feel works is to identify your strengths and relate them back to the specific domain, work on them and allow yourself to shine.

I think it's more about finding out what you're good at and what you enjoy most and deepening and expanding that part of yourself in that area and in your professional activities. This will significantly increase your chance to shine and to be happy.

As a senior woman now working in artificial intelligence, what advice would you give to your younger self?

I would advise my younger self to be more courageous, take the initiative and responsibility, and actively ask for feedback in order to improve. I would also recommend myself to make it a good habit to learn something new on a regular basis. Learning is a never-ending journey, and it will refresh our minds and keep us more active.

Handling the wide range of technical and business tasks sounds like a tough challenge - which is the best way for you to manage it?

The best way for me to do that is to set priorities and let go of the perfectionist in me. It's natural for all of us to want to get the best out of things, and that's where I find myself slowing down and dragging myself back. For instance, when I started a technical task, I wanted to understand every single detail of what was behind it, which was impossible and impractical. To be able to handle different types of tasks, quick success is very important. I prefer to spend 20% of the effort to get 80% of the work done and then give myself time to think about the draft and improve it.

There is a growing number of events where Women in Tech are in the spotlight. We've learned that you'll be appearing at one soon, too. Tell us more about it!

The Woman in Data Science (WiDS) initiative is from Stanford University in the US and aims to encourage women to get involved in the field of data science. Some high school students in Beijing, China, hosted it on June 5, 2022, and I participated in a panel discussion there to discuss professional development-related topics. You can watch the recording and visit their homepage for more details.

Thank you for the inspiring insights, Haodan! It was a great pleasure to get to know a new viewpoint, and we are sure that we were once again able to contribute to empowering women in technology. We're looking forward to the next part, where there will be another exciting perspective switch - stay tuned! 😉

Follow to not miss our upcoming interviews:

AI & ML SAP Community page

AI topic on SAP Community

AI Q&A on SAP Community

sdaho

Provide excellent customer service with SAP Service Ticket Intelligence

2022-08-16T15:20:28+02:00

Enterprises are facing challenges in customer service area.

While a typical customer journey includes a variety of interaction points, customer service interactions are among the most critical. They have the power to either strengthen customer relationships or lead to churn. A negative experience in customer service can have dreadful consequences for business. The customer service centers are facing challenges due to:

Increasing complexity and volume of tickets

Omni-channel customer engagement creates large volumes of unstructured text data

Lack of transparency and visibility into history of ticket resolution hinders informed decision making

Labor-intensive and slow resolution of customer service tickets

81% of all customers attempt to take care of matters themselves before reaching out to a live representative

The COVID-19 pandemic amplified the above challenges.

Service ticket intelligence Overview

As part of SAP AI Business Services, the SAP solution portfolio of intelligent services and applications, SAP created an intelligent business service to automatically categorize incoming service tickets and provide resolution recommendations for the agents. This allows contact centers to expedite problem resolution and scale with the increase of digital service requests. The speed of resolution is important because customers expect their complaints to be addressed quickly, and a 24-hour response window doesn’t necessarily match their expectations. For example, on a social media channel most people expect service responses within an hour.

To help companies keep up with customers’ growing expectations, SAP Service Ticket Intelligence application efficiently processes inbound social media posts, e-mails, and other channel interactions. It automatically categorizes customer tickets, proposes solutions and prioritizes the tickets for customer service employees.

With SAP Service Ticketing, incoming customers service tickets are automatically classified into their categories and routed to the right agent. The agent receives suggested solutions to response to customer accurately.

Service Ticket Intelligence typical user case system Demo

Customer sent a complaint

Service cloud creates a service ticket for this customer complaint.

Open the ticket.

The Incident Category is filled with the appropriate category implemented with machine learning. For the text received in the email, the service category is proposed as Complaint and the Incident category as Missed Appointment.

Furthermore, we can see that the ticket got identified as Urgent Priority, that the sentiment of the ticket is weak negative.

In addition to this, we can also see the “Similar Tickets Recommended”, in the similar tickets facet as shown below.

Further, we can directly provide a suitable solution by going to Interactions and clicking on Reply All.

We can see that template is auto assigned based on E-mail template recommendation scoring. Send the response.

Get hands-on skills from SAP Discovery Center

Try start a mission in SAP discovery center

User the link to access SAP discovery centre SAP Discovery Center (cloud.sap)

If you don’t have a user yet, you can sign up an account for free.

Search for “service ticket intelligence”.

Click the mission tile.

You can start the mission or if you already started the mission, click the top-right link to enter mission details.

Follow steps in the mission to get hands on experience of service ticket intelligence

Discover Basics section provides more informational details to let you know more about SAP discovery center and SAP AI business services. Feel free to explore links within the section as they are optional.

SETUP GUIDE section is a must-do in this mission.

Acquire BTP trial account and prepare POSTMAN

As a preparation for the following steps you have to create a free trial account on the SAP Cloud Platform by completing this tutorial which additionally covers details on the privileges available with Enterprise versus Trial Accounts.

To successfully test the model you have to use a tool called Postman. Please make sure to install and set it up by completing the tutorial which can be found here.

Create BTP service instance for service ticket intelligence

Once you have finished both tutorials your SAP Cloud Platform trial account should look like this:

To continue, you must enter your trial account. If this is your first time entering the Account you have to choose the region and other entitlements. In order to create a service instance for Service Ticket Intelligence please follow this tutorial.

Setting up Postman to Call Service Ticket Intelligence APIs

As your next step you must adjust the Postman environment.

In this tutorial you will import the Postman environment and the Postman collection in order to be able to test the ticket classification and the solution recommendation of service ticket intelligence.

Use Postman to Classify Service Requests

First you have to upload the training data and train your machine learning model. During the training, the machine learning model analyzes and recognizes patterns in your data. Based on these, the model is now able to make own predictions for new incoming data. Depending on the amount of data the training may take some time.

After your model is trained it needs to be activated in order to be able to make predictions.

By following this tutorial you will train and deploy a Service Ticket Intelligence machine learning model. Furthermore the tutorial results in testing the model with test data and receive the prediction of the ticket classification.

The classification prediction contains a confidence score. The higher the value of the confidence score of the prediction, the more reliable is the result. A sample response containing the confidence score can be found here.

Use Postman to Get Solution Recommendations

As you have learned in the previous mission you first will have to train and activate the machine learning model again.

This tutorial guides you through the steps from uploading the training data to being able to test the service and get solution recommendations.

Similar to the ticket classification response, the solution recommendation also contains a confidence score to show which solution may be the most relevant for the agent. A sample response can be found here.

Other available resources

You can click “resource” tab to access more resources available for service ticket intelligence.

Wrap Up

Business Benefits

Automated service ticket classification frees contact center agents from repetitive, tedious tasks, as they can now rely on a prioritized worklist of issues to address. As a result, they can spend more time on customer interactions, improving customer trust and satisfaction for a better business outcome.

Key Benefits:

Deliver satisfying customer services

Expedite service desk response time

Implement flexibility for simple user experiences

Higher rates of problem resolution and ticket closure •Simplified customer service processes

Improved ability to scale shared services as the business grows due to an increase in digital interactions

For more details regarding this service, please visit Service Catalog.

Additional

Please share your feedback or any thoughts in a comment

You can follow SAP AI Business Service Topic page. It is a great community and you can post and answer questions and read other posts on the topic.

Please follow my profile for similar contents on SAP Customer Experience related solutons.

Uploading large training files with Service Ticket Intelligence

2022-08-23T12:52:21+02:00

For machine learning to accurately understand and curate actions that caters to business requirements, the quality and quantity of training data are key ingredients for training highly efficient and effective machine learning models.

Having large volumes of good, well represented and labelled data will inevitability translate to larger training file sizes. In the context of Service Ticket Intelligence, this becomes a problem given that the service only allows training file uploads of up to 75MB in size. Users are then limited by the number of records they can upload for model training or forced to make use of the more complex OData protocol to upload larger training files.

In order to address this limitation, a new bulk file upload feature has been introduced with the 2107 release in Service Ticket Intelligence. Users can now perform a multi-part file upload operation for the model to learn with a larger training dataset. This can potentially result in a significant improvement with model performance.

Bulk file upload can be applied and used for all scenarios (classification, recommendation and clustering) supported by Service Ticket Intelligence.

Bulk File Upload Overview

Bulk File Upload Process

The process of uploading large files via bulk file upload will be as follows:

1. File splitting

Split the CSV file into multiple smaller CSV files.

Each file part should not exceed 75MB

Ensure that the column names and their respective data are present and consistent

2. Indicate bulk upload operation

Indicate in the file upload endpoint (/sti/training/model) that this operation is a multi-part file upload.

bulk_file_upload is set to true

bulk_file_upload_complete is set to false

Sample request body as follows:

{

   "scenario":{

      "desc":"Training Classification travel data with priority and keywords (small)",

      "type":"classification",

      "language":"en",

      "business_object":"ticket"

   },

   "options":{

      "bulk_file_upload":true,

      "bulk_file_upload_complete":false

   },

   "mapping":{

      "input":[

         "description"

      ],

      "output":[

         "category"

      ],

      "keywords":[

         {

            "columns":[

               "description"

            ],

            "text":[

               "Name",

               "Surname"

            ],

            "value":"1"

         }

      ]

   },

   "training":{

      "file":" c3ViamVjdCxib2R5LGFydGljbGVfaWQsYXJ0aWNsZV9kZXNjLGRhdGFzb3VyY2UsYXJ0aWNsZV91cmwKIkknbSBhIHRyaXBsZSBDYXByaWNvcm4gKFN1biwgTW9vbiBhbmQgYXNjZW5kYW50IGluIENhcHJpY29ybikgV2hhdCBkb2VzIHRoaXMgc2F5IGFib3V0IG1lPyIsLDEscmVsaWdpb3VzIHNlbnRpbWVudHMsbWluZHRvdWNoLHd3dy50aW1lc29maW5kaWEuY29tCldoYXQgc2hvdWxkIEkgZG8gdG8gYmUgYSBncmVhdCBnZW9sb2dpc3Q"

   }

}

3. Last file part?

If there are 2 file parts or more remaining to be uploaded, proceed to Step 4.

Else, skip to Step 5 if this is the final file part.

4. Indicate reference model id for subsequent upload

To upload the subsequent multi-part file, make a reference to the model id generated from the previous request in the file upload endpoint (/sti/training/model). Once done, repeat Step 3.

bulk_file_upload_reference_model is referenced

bulk_file_upload_complete is false

Sample request body as follows:

{

   "bulk_file_upload_reference_model_id":"90c6daa1894143dda96410bd1e1b70c7",

   "bulk_file_upload_complete":false,

   "training":{

      "file":“c3ViamVjdCxib2R5LGFydGljbGVfaWQsYXJ0aWNsZV9kZXNjLGRhdGFzb3VyY2UsYXJ0aWNsZV91cmwKIkknbSBhIHRyaXBsZSBDYXByaWNvcm4gKFN1biwgTW9vbiBhbmQgYXNjZW5kYW50IGluIENhcHJpY29ybikgV2hhdCBkb2VzIHRoaXMgc2F5IGFib3V0IG1lPyIsLDEscmVsaWdpb3VzIHNlbnRpbWVudHMsbWluZHRvdWNoLHd3dy50aW1lc29maW5kaWEuY29tCldoYXQgc2hvdWxkIEkgZG8gdG8gYmUgYSBncmVhdCBnZW9sb2dpc3Q”

   }

}

5. Indicate bulk file upload operation complete

For the final file part, indicate in the file upload endpoint (/sti/training/model) to complete the multi-part file upload.

bulk_file_upload_reference_model is referenced

bulk_file_upload_complete is true

Sample request body as follows:

{

   "bulk_file_upload_reference_model_id":"90c6daa1894143dda96410bd1e1b70c7",

   "bulk_file_upload_complete":true,

   "training":{

      "file":“c3ViamVjdCxib2R5LGFydGljbGVfaWQsYXJ0aWNsZV9kZXNjLGRhdGFzb3VyY2UsYXJ0aWNsZV91cmwKIkknbSBhIHRyaXBsZSBDYXByaWNvcm4gKFN1biwgTW9vbiBhbmQgYXNjZW5kYW50IGluIENhcHJpY29ybikgV2hhdCBkb2VzIHRoaXMgc2F5IGFib3V0IG1lPyIsLDEscmVsaWdpb3VzIHNlbnRpbWVudHMsbWluZHRvdWNoLHd3dy50aW1lc29maW5kaWEuY29tCldoYXQgc2hvdWxkIEkgZG8gdG8gYmUgYSBncmVhdCBnZW9sb2dpc3Q”

   }

}

6. Perform get model status

Perform a get model status call (/sti/training/model/status?model_id={{model_id}}).

Model status will transition from DATA_UPLOAD_IN_PROGRESS to DATA_PREPROCESSING_PENDING to NEW

7. Trigger model training

Model training can be triggered when model status is NEW.

Additional Resources

Postman collection: https://github.com/SAP-samples/service-ticket-intelligence-postman-collection

Bulk file upload API specs: https://help.sap.com/viewer/5088c3bb02144e7782959bb1529ca70e/SHIP/en-US/b0e6467e92c4455f83c0059b46fbe928.html

Bulk file upload examples: https://help.sap.com/viewer/5088c3bb02144e7782959bb1529ca70e/SHIP/en-US/f692293ec4ba4525b3755ef5f710dd3f.html

To try out other features (classification, recommendation, clustering scenarios) of Service Ticket Intelligence, we are also available on SAP BTP as a free trial.

Find out how you can set up your own trial account with Service Ticket Intelligence here and give the service a try today!

Mashup 3rd Party SMS Message Integration in SAP C4C

2023-06-01T10:49:52+02:00

In today's rapidly evolving world, mobile phones have emerged as indispensable gadgets, serving as the cornerstone of our daily lives. Despite the waning significance of Short Message Service (SMS), it remains a preferred communication channel for companies to connect with their customers. Furthermore, the transmission of crucial registration codes necessitates the utilization of SMS. Consequently, the integration of SMS functionality within SAP C4C assumes highly importance.

In this insightful blog post, we aim to guide you through the configuration process of your third-party SMS message integration. Additionally, we will provide you with a comprehensive tutorial on how to rigorously test your integration by employing manual messaging techniques within the SAP C4C environment.

In SAP C4C there are 3 steps to configure your SMS integration;

1- Enable the SMS Channel

2- Configure a Mashup SMS Service

3- Configure a SMS Channel

1- Enable the SMS Channel

Under Business Configuration --> Implementation Projects --> Edit Project Scope --> Editing Questions --> Sales --> Account and Activity Management --> Activity Management --> Enabling "Do you want to record messaging activity ?"

Please click to enlarge

2- Configure a Mashup SMS Service

Under Administrator --> Mashup Web Services --> New --> Rest Service --> Filling the areas needed

Please click to enlarge

Basically, one of the most significant issue is filling the "BODY" area. To sending SMS you need to fill this area correctly with the information from company and 3rd Party SMS Provider. Also for MESGBODY and NUMBERS areas need to fill SAP C4C Body logic.

Authentication Type: Depends on supplier and company. In this example it is basic.

Service Protocol: XML

HTTPS Method: POST

Also, after you may press simulate button on below to try your mashup.

3- Configure a SMS Channel

Configuring SMS channel is mandatory to sending SMS Messages in service tickets. In this configuration you need to name your configuration and select ticket type.

Under Administrator --> Service and Social --> Messaging Services --> New

Please click to enlarge

Testing SMS Manually

After all of these configurations, you may try your SMS Message services from tickets.

Under Tickets --> Select one of the ticket or create new --> Interactions Tab --> New --> Select SMS --> Choose your SMS Service --> Enter phone number to send SMS --> Enter text and send

Please click to enlarge

In conclusion, in today's dynamic world, mobile phones are indispensable, with SMS remaining a crucial communication channel for businesses to connect with customers. In this post you may learn how to make 3rd party SMS Integration for SAP C4C. Also, you may test your integration manually on ticket workcenter.

Stay tuned as we guide you through the process and empower you to enhance customer communication within SAP C4C.

Also you may find the SAP Video document for external SMS integration on this Microlearning Movie Link.

Finally, please do not hesitate to share your feedbacks for this post and do not forget to follow me from denizhanak

High Velocity and Business agility : Lean , Agile and ITIL

2023-06-16T06:57:57+02:00

I have invested some time in getting certified in few of the professional certifications . Some ware due to Organisational standards and some are out of interest . When I look at all those certifications one thing comes to my mind is there a way all these Practices , Processes and Tools can be combined to achieve higher Organisational Goals and success . Would merging these approaches assist the organisation, its staff, and its end users all at the same time? In this article, I'll discuss my thoughts on the possibilities.

The metaphor of development teams tossing apps over a wall to the operations group for deployment is frequently used to demonstrate how development and operations function in silos. The DevOps movement began in an attempt to break down these barriers between the development and operations teams. The DevOps movement is based on a culture of understanding the capabilities and restrictions of the other group (development or operations): "Delivering value to the business through software necessitates processes and coordination that frequently span multiple teams across complex systems, as well as developing and delivering software that is both quality and resilient."

Agile, Lean, and ITIL are all terms that have one thing in common: they make IT service desk collaboration, communication, and success easier.

While these frameworks and philosophies share some characteristics, it's crucial to grasp each term separately so that you can put them to use together.

What is Agile ?

The Agile methodology is a project management technique that emphasises constant collaboration and improvement by splitting the project into phases. Teams go through a planning, execution, and evaluation cycle.

What is Lean ?

Lean approach is a means of optimising your organisation's people, resources, effort, and energy towards delivering value for the customer. It is founded on two guiding principles: constant improvement and human dignity.

What is ITIL ?

The Information Technology Infrastructure Library is a collection of detailed practises for IT operations including IT service management and IT asset management that focus on aligning IT services with business demands.

Is it possible to combine these all ? What do we call the Methodology which combines techniques and principles of all these individual practices ?

We fundamentally understand Agile is used in Development environments and ITIL is for Operations . But when Development and Operations merge together to make a DevOps team the practices should also be merged for making sure none of these teams are working in silos and yielding maximum results as intended .

We can call it ITIL V4 introduced in 2019 . All of this sounds fantastic, but there is no room in your business for experimenting with approaches that may or may not work together for your specific aims. So, what's the catch?

LEAN, ITIL, AND AGILE BENEFITS :

Faster issue resolution

Improved productivity

A better overall customer experience

Reduction in wasted time and ultimately, money

Contentious improvement

These advantages appear to be fantastic, but how can you encourage your staff to adopt this mix of strategies? The following suggestions may be useful:

Consider these three ways in the context of a practical situation where these benefits can be realised. Assume you work in the financial business, such as a bank or a credit union, and your IT service desk is in charge of ensuring that customers can access their money and workers can do their tasks. That's a lot of strain! Consider that consumers are unable to access their online banking.

If our team is completely reliant on ITIL, they may be stymied while attempting to handle events through typical incident management techniques. However, by combining the Agile methodology, which encourages multiple people to work on the project with constant feedback loops, and Lean IT, which encourages the use of self-service and slimmed-down processes, you can expect the problem to be solved much more quickly and without wasting time or money looking for a solution. This means your consumers can access their funds, which improves their experience, and your personnel aren't scrambling to deal with dissatisfied customers and devise their own solutions.

There are many more such business use cases found across various organisations . Now the question is how to harness all these goldmine of benefits ?

We discussed how to put these ideas into action previously, but when employing various frameworks or ideologies, it's critical to create an organised structure to connect them all together. You can accomplish this by utilising an IT service management platform.

Thank you for taking time to read this article.