A Survey on Machine Learning Approach for Stock Market Prediction

Stock market domain is a promising domain in Machine learning approach. Internet technologies helps to gathered various kinds of structured and unstructured data, such as blogs, message threads, an enormous amount of technical data. company meetings, quarterly results of company etc has been accumulated. Currently, many applications have used data mining techniques or machine learning to exploit such data. This study is a comprehensive overview on Machine Learning. Unlike other domains Stock market data has a correlation among different features, and its characteristics change variously with time, political and natural changes. In this paper, we discuss a various methodology for predicting power to manipulate stock prices. However, there is no structured defined frame for solving the problem in stock market domain. We believe the existing ambiguity on the topic is due to its interdisciplinary properties that require both factors affecting the economics as well as artificial intelligence. We review the work related to prediction of stock market based on text-mining and gives an outer picture of the generalized modules. Our comparative analysis expands the theoretical and technical aspects behind each.