Sentiment analysis method based on improved SO-PMI algorithm and fused word vector

The invention discloses a sentiment analysis method based on an improved SO-PMI algorithm and a fused word vector, and the method comprises the steps: constructing a domain sentiment dictionary which comprises an evaluation object dictionary, a negative dictionary, a degree adverb dictionary, a positive sentiment dictionary and a negative sentiment dictionary, the positive emotion dictionary and the negative emotion dictionary are calculated and determined through an improved SO-PMI algorithm; using a word2vec model to train the text in the corpus to generate an original word vector; dividing sentiment words into five dimensions by using the constructed sentiment dictionary, and constructing sentiment vectors; and fusing the original word vector and the emotion vector to generate a word vector fused with emotion information. According to the method, sentiment vectorization representation is carried out on a plurality of constructed dictionaries containing sentiment tendency and strength, the dictionaries are