Comparative Analysis of Recent Neuro Fuzzy Systems For Stock Market Prediction

The fuzzy logic and neural networks are the two powerful techniques commonly used in various application field from their evolution. The combination of both techniques is commonly termed as Neuro-Fuzzy System (NFS). Both techniques are combined because they overcome the limitations of each other. Basically, this model is used to construct the complex model by using fuzzy logic and its capabilities are improves with a neural network. In NFS, fuzzy rules are adjusted by the input-output patterns of a neural network. This paper presents the use of this powerful NFS system in the fields of stock market application for stock price prediction. Most of the traditional approaches do not consider all kind of stock price movements. But literature proves that the NFS is a leading technique in stock market prediction. This paper initially describes the brief description of neural networks and Fuzzy logic system along with their pros and cons. Further discussing, how the advantages of both systems combined in NFS. Also, the different types and architectures of NFS are presented here. Finally, this paper studies the recent NFS dependent methods which is used for the prediction of stock market. This paper summarizes the analysis of these techniques based on the technique used, the dataset used, their advantages and some research gap of all these techniques.