Comparative Study of Stock Price Prediction using Machine Learning

Kadu, P. P.; Bamnote, G. R.

The stock market is mainly an aggregation of different sellers and buyers of stock. 'A stock (also known as shares more commonly) in general represents ownership claims on business by a particular individual or a group of people'. The effort to find out the upcoming stock market value is recognized as a stock market prediction. The forecast is anticipated to be efficient, accurate and robust. The system should function based on the real-life circumstances and it have to be well-matched to realistic surroundings. The system is also expected to consider the entire constraints, which may have an effect on the stock's value and performance. In stock market the selection of stock is very important for trading and investment, if fundamentally the stock is good but if the sector of that stock is down then ultimately the one will lose its trade for buy position. That's why there is need to continuously know the updated knowledge of current situation, News that has direct or indirect impact on stock or sector. The recent example is due to covid-19 pandemic the auto sector was underperforming as the sell-value of vehicles are dropped down and the pharmacy sector is performing well due to covid-19 medicine. During past years many researchers have given contribution in this field but there is still a need to do the research for stock selection based on fundamental and technical analysis, this provides strong motivation for a system that can efficiently extract data from different web sources can be done using web crawling and capable of prediction of stock price, which would be helpful for individual for selection of stock for trading and investment. Research gaps and challenges between existing techniques are listed and detailed, which helps researchers to upgrade future works. Here, this paper provides an overview of the applications of machine learning in stock market forecasting to determine what can be done in the future. © 2021 IEEE.