Sensitive detection method and apparatus text

2014 
The present invention discloses a method and apparatus for detecting sensitive text belongs to the field of information technology process. A method comprising: obtaining a characteristic detecting the current text of the text string; feature text string automatically detected based on finite state machine pre-established, resulting in the characteristic frequency of each keyword appears in the text string; keyword category for a plurality of each category keywords, each keyword based on the keyword category corresponding to each keyword and the frequency of occurrence of the preset weight, calculated in the text keyword category weights weight; weight when the at least one keyword in the category of major when the predetermined threshold value, the text determined as sensitive text. When detecting features of the present invention, the text string, only once from beginning to end in accordance with the scanning finite state machine automatically pre-established, the detection efficiency is improved, to speed up the detection rate; and when determining sensitive text, based on the need to keywords preset weights, so improving the detection granularity.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []