Is supposed to be a tool for stock predicitons.<br > ###So far consists of three logical parts:

• scraper /* scrapes webpages and tries to parse them with regards to publication date, title, article */
• stock-value-importer /* takes a csv file from http://www.nasdaqomxnordic.com/ and imports it into the database */
• stockbrowser /* simplifies browsing articles related in time to stock value deviations */
• stock-predictor
  • word-frequency-predictor /* Tries to find any particular words that might occur more often in relation stock value deviations */

So far I have scraped all articles on Affärsvärlden, and tried to do a word-frequency analysis related to the ERIC_B stock. Yielding 0 prediction accuracy.. =) Who thought it would be that simple eh?<br > Next I will start looking into more sophisticated prediction techniques, basically inputing whole articles into a neural net and see what comes out the other end.

A database dump of all those articles resides here: https://github.com/jderehag/stockrobot/releases/download/v0.1/webpages.db.bz2

###Beautification disclaimer: Admittedly, this is not the prettiest of applications. This is very much a work in progress and main goal has really been to test various datamining and prediction techniques. Therefore the application is not strictly adherent to MVC pattern, and far from optimized.