Research on the Sentiment Analysis Based on Machine Learning and Feature Extraction Algorithm
Xiaofang Jin, Ying Xu · 2019
Sentiment analysis is a comprehensive research in Natural Language Processing (NLP) which is used to target the emotional context of the text. In previous experiments, shallow vocabulary and grammatical features were often considered, and the implicit semantic features were neglected. In order to solve this problem, this paper combines the training ways of the word vectors (Word2vec Doc2veC TF-IDF model) and machine classification model (SVM, I<; NN, etc.) for classification and then extracting features. The main goal is to use the performance of Word2vec Doc2vec and TF-IDF models to convert words into high-dimensional vectors. Through the calculation of the cosine similarity of the word vectors, the semantic similarity of the vocabulary can be obtained. Then these word vectors are used as semantic features and input into the machine learning classification model for training, so as to obtain the emotion classification results. In this paper, compared with the results of dictionary-based text sentiment analysis, the accuracy of machine learning methods are higher.