Network brief comment sentiment mining method

Abstract

The invention relates to a network brief comment sentiment mining method, which includes the following steps: (Step 1) loop control parameters and an sentiment dictionary which are related to the method are set; (Step 2) the sentiment dictionary is utilized to carry out sentiment polarity prior-processing on a brief comment data set: if a word in the brief comment data set appears in the sentiment dictionary, a sentiment polarity value defined by the sentiment dictionary is assigned to the sentiment polarity of the word; (Step 3) the sentiment polarity of the word of the prior-processed brief comment data set and subject preference are initialized; (Step 4) a subject-sentiment mixed model is utilized to constantly iteratively update Temp theta, theta dsl<k>, mu. lk<v> and pi d<l>; (Step 5) by means of pi d<l>, the sentiment polarity of each brief comment d is judged: for the brief comment d, if pi d<l1> is greater than pi d<l2> (wherein l1 is positive sentiment, and l2 is negative sentiment), the sentiment polarity of the brief comment d is judged as positive sentiment, otherwise the sentiment polarity of the brief comment d is judged as negative sentiment. The method can effective mine the true sentiments and viewpoints of users hidden in brief comments.

Description

technical field [0001] The invention relates to the technical field of network public opinion analysis, in particular to a method for excavating sentiments of network short comments applied to social networks under the Web2.0 environment. Background technique [0002] Web 2.0, whose core spirit is freedom, openness and sharing, makes users the protagonists of the Internet. Platforms such as social networking sites, microblogs, and BBS forums provide economical and convenient channels for netizens to express their opinions and exchange emotions. Generally speaking, users’ comments on these platforms are short but full of personal emotions and subjective tendencies. For example, different readers hold different views on the same news event, and different users have different opinions about a certain mobile phone. Personalized user experience, different movie lovers will leave different viewing comments for the same movie, and so on. Research on how to efficiently mine the opi...

AI Technical Summary

Problems solved by technology

From the above modeling process, it can be seen that although JST, S-LDA and ASUM can obtain the topic and sentiment polarity of each word (sentence) in the comment, for each sentence in the comment, the short comment-sentiment-topic distribution are the same (for example: the phone in the first sentence and the photos in the second sentence have the same probability of belonging to the topic "mobile phone"), which may cause the topics of these two words to not be captured correctly, because according to the network The writing habit of short comments, if the subject of the first sentence is the overall quality of the mobile phone, then the probability that the subject of the second sentence is still the overall quality of the mobile phone will be relatively low

Also, the method used by AUSM to capture sentence topics and emotional polarity may lead to failure to capture the true emotions of sentences with positive words and negative words coexisting, for example, for sentences containing emotional positive words (clear) and emotional negative words (hard) in MR In the second sentence, AUSM may not be able to correctly judge the emotional polarity of the sentence

In addition, the modeling process of S-LDA adopts the method of first judging the topic of words and then determining the emotional polarity of words, which is not in line with the behavior habits of web users for short comments.

[0005]In short, the existing sentiment analysis technology based on the LDA sentiment topic model does not take into account the user's behavior habits when writing short comments, but simply assumes that all the sentences in the short comments The topic probability distributions of all the topics are the same, which is contrary to the behavior habits of web users' short comments, which deviates from the user's real emotion to varying degrees, so it is not suitable for digging out the user's real emotion from online short comments

Images