An Approach To Textual Analysis For Stocks

Konstantopoulos, Theodoros

Student Work

An Approach To Textual Analysis For Stocks Public

Numerous scientific attempts have been made to predict the stock market but there has not been one method that accurately predicts stock price movement. A group of websites was chosen based on the Alexa site list for the investing sub-category located under the business category. A custom crawler was written for every site that is listed on the Alexa website. Weights were generated based on the information for total sites linking a website and a sentiment analysis was performed by an analyzer that is trained using social media and news data.

Creator