A method for sentiment analysis on a security research report using big data. The method comprises: receiving an input security research report to be analyzed (S01); performing sentence segmentation on said security research report, so as to obtain segmented sentences (S02); performing word segmentation on the segmented sentences to obtain segmented words (S03); determining a predictive segmented sentence by using a sentiment dictionary based on the segmented words (S04); determining sentiment types of the segmented words in the predictive segmented sentence according to the sentiment dictionary (S05); scoring the predictive segmented sentences based on the sentiment types of the segmented words and by using pre-set scoring rules (S06); calculating, based on scoring calculation, the overall sentiment score of said security research report (S07); and according to a result of comparison between the overall sentiment score of said security research report and a pre-set score threshold value, acquiring a sentiment analysis result of said security research report (S08). The sentiment analysis on a security research report is implemented using a big data analysis and intelligent scoring method, which can solve the problem of low efficiency and accuracy of a sentiment analysis solution for a security research report in the prior art, and improve the efficiency and accuracy of sentiment analysis on the security research report.
展开▼
