A Modified Deep Learning Enthused Adversarial Network Model to Predict Financial Fluctuations in Stock Market

Abstract

Predicting financial fluctuations in the real-time stock market is considered to be a major problem due to dynamic changes in financial data. With the advent of using artificial intelligent techniques in the context of predicting the patterns, artificial neural networks have drawn the attention of various researchers to implement the same in several computational applications. Addressing this problem, a modified adversarial network based framework is proposed with the integration of gated recurrent unit and convolution neural network. The main objective of this model is to acquire data from online financial sites and to process the obtained information using adversarial network to generate predictions.