{
  "$schema": "/api/schemas/documents.schema.json",
  "apiVersion": "1.0.0",
  "generatedAt": "2026-07-14T19:46:21.019Z",
  "sourcePath": "src/main/data/docs",
  "docId": "10.5594-JMI.2025-OJUX5368",
  "document": {
    "abbrevTitle": "SMPTE Motion Imaging Jour.",
    "abbrevTitle$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "abstract": "Broadcasting organizations produce large volumes of news articles daily, requiring accurate metadata to enable efficient reuse across television and online platforms. Manual annotation, however, is both time-consuming and laborintensive. To address this, we propose an AI-based system for automatic multilabel classification of news text. A central challenge in this task is the imbalanced label distribution, where high-frequency labels dominate and rare labels are underrepresented. To mitigate this, we introduce Weighted Asymmetric Loss (WASL) with Label Smoothing, which integrates class-balanced weighting, suppression of dominant negative samples, and smoothing based on label co-occurrence to improve performance on infrequent labels. Evaluation on Reuters-21578 and Japan Broadcasting Corp. (NHK) News Web datasets demonstrates that our approach significantly outperforms baseline methods on both macro-F1 and rnicro-F1 scores. We further developed a prototype system and deployed it in local NHK broadcasting stations, where it reduced metadata creation costs and facilitated content reuse, while high-lighting practical considerations for workflow adaptation.",
    "abstract$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "authors": [
      {
        "name": "Yuki Yasuda"
      },
      {
        "name": "Simon Clippingdale"
      },
      {
        "name": "Taro Miyazaki"
      },
      {
        "name": "Jun Goto"
      },
      {
        "name": "Takahiro Mochizuki"
      }
    ],
    "authors$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "contentType": "orig-research",
    "contentType$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "copyright": {
      "holder": "SMPTE",
      "holder$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      },
      "year": "2025",
      "year$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "copyright$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docId": "10.5594-JMI.2025-OJUX5368",
    "docId$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docLabel": "MIJ 2025, Volume 134, Number 6 (pp. 16 to 24)",
    "docLabel$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docTitle": "Multi-Label Indexing Technology for News with AI-Based Text Processing",
    "docTitle$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docType": "Journal Article",
    "docType$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "doi": "10.5594/JMI.2025/OJUX5368",
    "doi$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "href": "https://doi.org/10.5594/JMI.2025/OJUX5368",
    "href$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "issn": {
      "electronic": "2160-2492",
      "electronic$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      },
      "print": "1545-0279",
      "print$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "issn$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "journalAcronym": "MIJ",
    "journalAcronym$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "journalTitle": "SMPTE Motion Imaging Journal",
    "journalTitle$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "keywords": [
      "Natural Language Processing",
      "Multi-Label Text Classification",
      "Loss Function",
      "Few-Shot Learning"
    ],
    "keywords$meta": {
      "confidence": "high",
      "note": "Long-tail keyword cleanup (keywordLongTailApply.js): prose/contentless terms dropped, acronym casing + typos fixed, duplicate variants folded; all surviving terms indexed in controlledKeywords.",
      "source": "resolved",
      "updated": "2026-07-14T16:29:38.680Z"
    },
    "number": "6",
    "number$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "pages": "16–24",
    "pages$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "publicationDate": "2025-10-01",
    "publicationDate$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "publisher": "SMPTE",
    "publisher$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "status": {
      "active": true,
      "active$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "volume": "134",
    "volume$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    }
  }
}