Sentence sentiment classification system and method based on sentiment dictionary construction by the price fluctuation and convolutional neural network

摘要

The present invention relates to a sentence emotion classification system and a method based on emotional dictionary construction and a composite-product neural network based on price fluctuations. The sentence emotion classification system according to the present invention includes a data preprocessor, an emotion dictionary construction unit, and an emotion classification unit. The data preprocessor extracts a sentence from a document set, analyzes morphemes, and removes words without discrimination from the document set to process stopwords. The emotional dictionary construction unit extracts a word and expresses it as a vector by using internal / external boundary values of the word in order to determine the meaning and frequency of the word in the document set, collects price fluctuation information according to a specific field, and the price fluctuations. The emotional dictionary is created by classifying the positive and negative of the words based on the information. The emotion classification unit classifies the emotions of sentences based on the emotional dictionary using a convolutional neural network (CNN).