{
  "deformat_commit": "63d89f5db77bcbf704b41ad0690199e237feae72",
  "dataset": {
    "name": "murrough-foley/web-content-extraction-benchmark",
    "revision": "be72432cfa012ac918af47010bf106a2801afeef",
    "split": "dev",
    "pages": 1497
  },
  "commands": [
    "scripts/fetch_wcxb.py --split dev",
    "cargo run --release --example bench_wcxb -- --split dev --extractor strip",
    "cargo run --release --example bench_wcxb -- --split dev --extractor triple",
    "cargo run --release --example bench_wcxb -- --split dev --extractor both"
  ],
  "results": {
    "strip": {
      "article": {"n": 793, "f1": 0.854, "precision": 0.773, "recall": 0.992},
      "collection": {"n": 117, "f1": 0.522, "precision": 0.405, "recall": 0.934},
      "documentation": {"n": 91, "f1": 0.911, "precision": 0.864, "recall": 0.986},
      "forum": {"n": 113, "f1": 0.512, "precision": 0.429, "recall": 0.766},
      "listing": {"n": 99, "f1": 0.612, "precision": 0.522, "recall": 0.904},
      "product": {"n": 119, "f1": 0.450, "precision": 0.330, "recall": 0.937},
      "service": {"n": 165, "f1": 0.748, "precision": 0.654, "recall": 0.967},
      "all": {"n": 1497, "f1": 0.746, "precision": 0.659, "recall": 0.957}
    },
    "triple": {
      "article": {"n": 793, "f1": 0.880, "precision": 0.826, "recall": 0.972},
      "collection": {"n": 117, "f1": 0.551, "precision": 0.501, "recall": 0.794},
      "documentation": {"n": 91, "f1": 0.904, "precision": 0.901, "recall": 0.933},
      "forum": {"n": 113, "f1": 0.561, "precision": 0.502, "recall": 0.735},
      "listing": {"n": 99, "f1": 0.620, "precision": 0.599, "recall": 0.776},
      "product": {"n": 119, "f1": 0.500, "precision": 0.401, "recall": 0.853},
      "service": {"n": 165, "f1": 0.788, "precision": 0.743, "recall": 0.911},
      "all": {"n": 1497, "f1": 0.774, "precision": 0.723, "recall": 0.909}
    },
    "both": {
      "article": {"n": 793, "f1": 0.880, "precision": 0.820, "recall": 0.980},
      "collection": {"n": 117, "f1": 0.558, "precision": 0.498, "recall": 0.820},
      "documentation": {"n": 91, "f1": 0.913, "precision": 0.898, "recall": 0.953},
      "forum": {"n": 113, "f1": 0.560, "precision": 0.497, "recall": 0.744},
      "listing": {"n": 99, "f1": 0.619, "precision": 0.593, "recall": 0.787},
      "product": {"n": 119, "f1": 0.496, "precision": 0.394, "recall": 0.864},
      "service": {"n": 165, "f1": 0.791, "precision": 0.738, "recall": 0.926},
      "all": {"n": 1497, "f1": 0.775, "precision": 0.717, "recall": 0.920}
    }
  }
}
