Term vector-based abbreviation ambiguity elimination system and method

2015 
The invention relates to a term vector-based abbreviation ambiguity elimination system and method. The method comprises the steps of removing all non-alphabetic symbols and stop words in a file to be detected containing target abbreviations; taking words which appear before and after the target abbreviations within a fixed length as alternative keywords, screening the alternative keywords according to relative importance to obtain keywords-in-context, and gathering the keywords-in-context of all the target abbreviations; conducting summation on term vectors in a term vector set and term vectors corresponding to the keywords-in-context, so that term vector representation of each target abbreviation in the file to be detected is obtained; finally, comparing the term vector representation of the target abbreviations with the term vector of each meaning of each target abbreviation in an abbreviation bank, and taking the most similar meaning as the meaning of each target abbreviation in the file to be detected. Therefore, the system and method can be widely applied to the field of abbreviation ambiguity elimination.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    2
    References
    0
    Citations
    NaN
    KQI
    []