|
Abstract:
|
We analyzed the stock market index based upon the market capitalizations of 500 large companies in the US (this historical data contained percentage returns for the S&P 500 stock index). The objective of this analysis is firstly to predict market moves. This prediction is largely based on Yahoo historical price data using neural networks. Secondly, we have used sentiment analysis to view the percentage and polarity of recent tweets concerning stock symbols. Stock moves prediction and sentiment analysis of stock tweets is thus combined in order to maximize the accuracy, which regards choosing the similar results of both methods by association, then executing an effective trading strategy of buying and selling the concerned stock at just the right moment. Finally, we implemented this previously described trading strategy in a web application that you will be presented.
|
