The following work is regarding the prediction of stocks with the techniques used in sentiment analysis. First of all a large number of articles are downloaded then cleaned and formatted with techniques from Data Mining.
In the second step 3 Lexicons will be used for sentiment analysis, first two are L&M and Lui, the third one is a trained lexicon based on pre-labelled datasets.
The third step is about predicting the stocks, for this purpose several models will be created and used on specific stocks.