Analysis of AI in stock market based on natural language model

Abstract

Stock is a high-risk and high-yield way of investment and financial management, which occupies an important position in the national economy, and requires investors to have rich knowledge and experience in finance. With the development of the Internet and artificial intelligence, it has attracted many investors, and the integration of artificial intelligence and the financial market has changed the traditional way of securities trading. It has become a popular phenomenon for investors to express their stock opinions on stock forums, which produces a lot of stock comments, and the investor sentiment contained in this information often affects the trend of the stock market. This paper through python web crawler technology, climb Weibo related topic text, using the characteristics of a large number of text semantic expression, on the basis of traditional machine learning methods, through the in-depth study of word frequency-inverse document frequency (TF-IDF) feature extraction method, realize the category of multiple comment corpus text emotion accurate analysis, we found that the public for AI application in the stock market in anticipation at the same time also have some doubts. To provide a certain development direction for artificial intelligence in the future stock market.