An Amalgamation of Big Data Analytics with Tweet Feeds for Stock Market Trend Anticipating Systems: A Review

In digital era, the provision of streaming data and its storage has taken wider forms where data can be in structure, semi-structured and unstructured forms. As data is increasing the storage capacity also has to be increased, and the processing of data from such huge storage may be time consuming. And it is tricky to handle and process such data via conventional software and database procedures, which lead to the research toward big data analytics. It aims at handling of massive data storage with fast processing techniques and to help companies in optimizing business, advancement of operations, making more intelligent, and fast decisions. Hence Big data analytics is an important field that derives insights from the data and prediction system is one of the famous applications of it. It takes the historical data and analyses, and then it forecasts the past and future situations basing on identified hidden patterns of data considered. The analytics can be categorized into different forms namely Business Analytics (BA) and Predictive Analytics (PA). The inclusion of skills, technologies, applications, and processes with statistical techniques adopted by organizations for their data available to impel business planning is referred to as business analytics. The forecasting approach to foresee upcoming events and trends known as Predictive Analytics, which identifies the hidden patterns and determines what is likely to happen from the historical information available using statistical and mathematical models. To enhance the forecasting process, an opinion mining can also be included. Nowadays sentiment of the people also considered to improve the accuracy level of anticipation. Effecting factors should be considered and clearly analyzed to construct accurate model so as to supply most relevant suggestions. Several researchers proposed various prediction algorithms and methods in order to construct the accuracy improved model and user satisfaction. In this chapter, authors studied various anticipating models and discussed their preference criteria. As a part of that, we studied various important preference factors in stock trend prediction and categorized them based on effecting factors. This chapter reports prospect directions in prediction models and compiling an easy guide reference list to help out the researchers.