Research Area:  Machine Learning

Abstract:

Stock market variation data is collected in one form of breaking news from various finance web sites. The free financial data about companies is available on Internet Portals .The stock market sentiment changes with at a fraction due to major financial reforms, weather, natural disasters, and news events. Online finance news generates large amount of data. The market reforms are predicted with various machine learning algorithms. The term frequency-inverse document frequency (TF-IDF) features extracted from online news data for various companies of Bombay Stock Exchange are used along with other stock market features for prediction. The next day-s stock price is predicted using ensemble deep learning framework. The data set is optimized by various deep learning techniques to get more accurate results. The proposed model produces approximately 85% of accurate prediction with deep learning framework. The market trends in terms of high and low stock values are matching exactly. The results can be further improved with use of high frequency trading algorithms.

Keywords:  

Author(s) Name:  Vaishali Ingle, Sachin Deshmukh

Publisher name:  ELSEVIER

DOI:  10.1016/j.gltp.2021.01.008

Volume Information:  Volume 2, Issue 1, June 2021, Pages 47-66

Paper Link:   https://www.sciencedirect.com/science/article/pii/S2666285X2100008X