International Journal of Electrical and Computer Engineering (IJECE) Vol. 6, No. 6, December 2016, pp. 3196~3204 ISSN: 2088-8708, DOI: 10.11591/ijece.v6i6.13323  3196 Journal homepage: http://iaesjournal.com/online/index.php/IJECE Performance Forecasting of Share Market using Machine Learning Techniques: A Review Sachin Kamley 1 , Shailesh Jaloree 2 , R. S. Thakur 3 1 Department of Computer Applications, S.A.T.I., Vidisha, India 2 Department of Applied Math’s & Computer Science, S.A.T.I., Vidisha, India 3 Department of Computer Applications, M.A.N.I.T., Bhopal, India Article Info ABSTRACT Article history: Received Aug 20, 2016 Revised Nov 12, 2016 Accepted Nov 26, 2016 Forecasting share performance becomes more challenging issue due to the enormous amount of valuable trading data stored in the stock database. Currently, existing forecasting methods are insufficient to analyze the share performance accurately. There are two main reasons for that: First, the study of existing forecasting methods is still insufficient to identify the most suitable methods for share price prediction. Second, the lack of investigations made on the factors affecting the share performance. In this regard, this study presents a systematic review of the last fifteen years on various machine learning techniques in order to analyze share performance accurately. The only objective of this study is to provide an overview of the machine learning techniques that have been used to forecast share performance. This paper also highlights a how the prediction algorithms can be used to identify the most important variables in a share market dataset. Finally, we could have succeeded to analyze share performance effectively. It could bring benefits and impacts to researchers, society, brokers and financial analysts. Keyword: Machine learning Performance forecasting Share market Copyright © 2016 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Sachin Kamley, Departement of Computer Applications, S.A.T.I., B.T.I. Road, Sherpura, Vidisha, 464001, MP. India. Email:skamley@gmail.com 1. INTRODUCTION Now a day, share price prediction is an important concern for policy makers, researchers and investors because accurate price prediction plays key role in investment decision making. In general, stock market nature is considered to be chaotic and complicated, but it has been influenced by several economic and external environmental factors. Therefore, share market analyses have been using some approaches for predicting share prices. The random walk theory states that share price movements are independent of each other and price movements do not follow any patterns or trends [1]. Thus, it is practically impossible to predict the future price movements based on the historical data. On the other hand, technical analysis can be used to identify the patterns and trends based on the historic prices [1]. So therefore, future price movements can be done by examining past share prices. For many years, technical analyses with statistical approaches have been widely applied to this area in order to develop some concepts and strategies to be helpful for share performance forecasting [1-2]. Currently, there are various techniques have been proposed to evaluate share performance. Machine learning is one of the core areas which has been widely used to analyze share performance. The main objective of machine learning techniques is to automatically learn and recognize patterns from huge amounts