We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
STOCK PRICE PREDICTION BASED ON FINANCE RELATED NEWS USING NLP, LASSO AND ARIMAX.
- Authors
SUDHAKAR, KALVA; NAGANJANEYULU, S.
- Abstract
Stock trends in the finance sector are particularly critical and unpredictable in nature. In recent years, it has attracted the attention of researchers to study the patterns on its uncertainty for better forecasting. Stock trends are used in order to help investors and industry analysts evaluate market behaviour and prepare their investment strategies accordingly. There are many influences driving stock trends, one of which is daily news releases. In order to allow investors to make good trading decisions before making actual investments, it is important to thoroughly predict the impact. The aim of this paper is to merge the traditional numerical stock prediction model which is designed by using LASSO and ARIMAX with related financial news analysis. To process the huge amount of news dataset Mahout and map reduce estimating algorithm has been used, and executed on Hadoop to get faster results. Our results show that the learning model based on numerical features and NLP features using historical prices has shown the best performance.
- Subjects
STOCK price forecasting; FINANCIAL services industry; INVESTMENTS; INVESTORS; BUSINESS analysts
- Publication
i-manager's Journal on Software Engineering, 2020, Vol 14, Issue 4, p11
- ISSN
0973-5151
- Publication type
Article
- DOI
10.26634/jse.14.4.17661