... the60 million token corpus, Abbreviations and short-hand were expanded, for example defib expandsto defibrillator. Table 3 shows some unknown to-kens and their resolutions. The proofreading re-quire ... semantic classes,word qualifiers, phrases, and parses the text usingits own grammar, and maps phrases to standardmedical vocabularies for clinical findings and dis-ease. The MetaMap (Aronson, 2001) ... abbreviations and shorthand were obtained from the hospital, and were manually compiled to resolve the meaning.Every alphabetic token was verified against thedictionary list, and classified into...