back to list

Config PubMed

Configuration for PubMed articles which are processed by PubAnnotation.
Creator jdkim@dbcls.rois.ac.jp
Body
{
  "break_pattern": "([ \t]*\n+)+[ \t]*",
  "candidate_pattern": "[ \t]+",
  "positive_rules": [
    [
      "[.!?]",
      "[0-9A-Z]"
    ],
    [
      "[:]",
      "[0-9]"
    ],
    [
      "[:]",
      "[A-Z][ a-z]"
    ]
  ],
  "negative_rules": [
    [
      "(Mrs|Mmes|Mr|Messrs|Ms|Prof|Dr|Drs|Rev|Hon|Sen|St)\\.",
      "[A-Z][a-z]"
    ],
    [
      "(Sr|Jr)\\.",
      "[A-Z][a-z]"
    ],
    [
      "\b[A-Z][a-z]*\\.",
      "[0-9A-Z]"
    ],
    [
      "(cf|vs)\\.",
      ""
    ],
    [
      "e\\.g\\.",
      ""
    ],
    [
      "i\\.e\\.",
      ""
    ],
    [
      "(Sec|Chap|Fig|fig|Eq)\\.",
      "[0-9A-Z]"
    ],
    [
      "\b[A-Z]\\.",
      ""
    ]
  ]
}