Writing feature and sequence feature combined Chinese sentiment new word recognition method and system

Abstract

The invention discloses a writing feature and sequence feature combined Chinese sentiment new word recognition method and system. The method comprises the following steps: for input text clauses, expressing the text clauses as sequences of various features (such as word, part of speech and the like) on the basis of writing features of writers of sentiment words and sequence features of the sentiment words; in allusion to the feature expressed text clauses, outputting sentiment word label sequences corresponding to the text clauses by utilizing a linear chain condition random field model, wherein the linear chain condition random field model is obtained on the basis of a traditional sentiment word-containing text through training; on the basis of sequences of words and label sequences of the sentiment words in the text clauses, recognizing the sentiment words in the text clauses by utilizing a finite state machine so as to form a sentiment word set; and finally filtering the sentiment word set by utilizing a Chinese old word bank, and taking the sentiment words which do not appear in the Chinese old word bank as Chinese sentiment new words. Through the embodiments of the invention, the technical problem of how to improve the sentiment new word recognition precision and the recall rate is solved.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of computer science, and in particular, to a method and system for recognizing Chinese sentimental new words combining writing features and sequence features. Background technique [0002] Text-oriented sentiment analysis has very important applications in market decision-making, public opinion analysis and other fields. As an important factor affecting the effect of sentiment analysis, sentiment words emerge in an endless stream over time. Therefore, automatic recognition of sentimental new words in text is of great significance for text sentiment analysis. With the advent of the self-media era, the massive social media texts accumulated on the Internet not only bring data support to the work of emotional new word recognition, but also pose severe technical challenges. [0003] The previous work on Chinese sentimental new word recognition can be divided into two categories: one of th...

AI Technical Summary

Problems solved by technology

In the past, the recognition methods of Chinese emotional new words mainly have the following shortcomings: (1) The method based on new word discovery needs to manually set and adjust the parameter threshold when discovering new words, which is not conducive to expansion and low efficiency; (2) The method based on new word discovery Often by filtering low-frequency new words to ensure accuracy, it is difficult to identify low-frequency new emotional words; (3) The method based on the context matching mode of emotional words only uses the limited characteristics of emotional words such as context vocabulary, part of speech, and syntactic structure, and ignores the context of words in sentences. Important information such as the location of the sentence, the punctuation mark of the sentence, the Chinese pinyin of the word, and the writing characteristics of the text author have restricted the recognition performance of emotional words.

Images