Progress in Machine Learning Techniques for Stock Market Movement Forecast

Abstract

Data-driven accurate stock market models can lead to timely, better decision making by the investors for a more profitable transaction. Such models can increase the chances of selecting more profitable stocks and reduce risk by avoiding risky investment. Last few decades of advances in soft-computing techniques in machine learning (ML), deep learning (DL), text mining (TM), and ensemble methods have positively reflected in the forecasting of stock market as well. In our work, we have reviewed some recent machine learning models for stock market forecasting. We have considered the works that cover various types for data sources, forecasting techniques, and efficient evaluation metrics. With our paper, we aim to provide a brief idea on the latest progress in stock market forecasting. We also summarize our analysis to highlight future research scopes in stock market movement forecasting.

Sentiment Analysis of Online Product Review using Deep Learning in Distributed Sensor Networks

“…In terms of current market analysis, sentiment analysis plays a vital role. Here, DSN transferred market-related data to machine learning and all the data was analyzed through linear algorithms that provide effective market forecasting to the organization [9]. On the other hand, through deep learning, an organization can analyze large amounts of valuable data in a fraction of a second through an automatic process.…”