{ "dataset": "party_statements", "name": "South Korean Party Statements", "description": "Official statements from party spokespersons and minutes from daily leadership meetings of South Korea's two major parties (Conservative and Progressive), covering 2003 to 2022. 83,201 total entries (35,115 conservative + 48,086 progressive). Parsed using the khaiii Korean morphological analyzer.", "time_coverage": "2003-2022", "data_version": "v2022", "package_version": "0.2.0", "variant": null, "default_variant": null, "available_variants": [], "variant_description": null, "recommended_use": null, "identifier_columns": "id", "text_columns": ["text", "filtered"], "supported_formats": ["csv", "parquet"], "managed_formats": ["csv", "parquet"], "artifacts": { "csv": { "format": "csv", "file": "sk_party_statements_v2022.csv", "download_url": "https://osf.io/download/8u2ah/", "sha256": "60874e7c44d851c9cfc0892d70f6ef9ff9fb3993a5324963297ca4eabd4868e4", "size_bytes": 740785920, "managed": true }, "parquet": { "format": "parquet", "file": "sk_party_statements_v2022.parquet", "download_url": "https://osf.io/download/8cjxu/", "sha256": "cee8a49adbe90f96ee4e2b45b6d84c433e5eb9ebb4849cfc979f6a19c57378ea", "size_bytes": 393216464, "managed": true } }, "columns": [ { "name": "no", "type": "integer", "description": "Sequential entry number within each party" }, { "name": "year", "type": "integer", "description": "Year the statement was posted" }, { "name": "ymd", "type": "character", "description": "Full date (YYYY-MM-DD)" }, { "name": "title", "type": "character", "description": "Title of the statement" }, { "name": "text", "type": "character", "description": "Full text of the statement" }, { "name": "filtered", "type": "character", "description": "Parsed text after morphological analysis; Korean-only" }, { "name": "partisan", "type": "character", "description": "Party affiliation: Progressive / Conservative" }, { "name": "conservative", "type": "integer", "description": "Binary indicator: 1 = Conservative Party, 0 = Progressive Party" }, { "name": "id", "type": "character", "description": "Unique document identifier (party prefix + entry number)", "identifier": true } ], "notes": { "missing_values": "Some fields may contain NA or empty strings.", "party_names": "Both parties have undergone frequent name changes. The 'partisan' column uses stable ideological labels rather than party names.", "text_processing": "All text is UTF-8 encoded Korean. 'text' contains the full original text; 'filtered' contains the morphologically parsed version." }, "extras": [] }