What is this website?
Casey Aggregate is a machine learning tool to understand the effect of media on financial markets. There are over one thousand daily newspapers in the United States, far too many for one person to read. An archive of the Wall Street Journal was crawled along with the change in the stock price over the next week. 1200 articles were entered into a machine-learning algorithm to 'teach' the computer the difference between good news and bad news. News articles from newspapers are fed into the algorithm to create a visual representation of the effect of the day's news on the stock market. This website currently aggregates from five newspapers: The Wall Street Journal, the New York Times, The LA Times, The New York Post, and USA Today. Hopefully, this project will be expanded to include many more news sources. The graphs show the Dow 500 (left) and the Russel 3000 (right). the Dow 500 is further divided into sectors (below). Individual judgments are often wrong, I am still working on the algorithm, but the graphic should be correct in aggregate.