The categorized financial news has been used significantly in stock prediction. Therefore, PSX-related financial news is categorized into the market, sector, and stock categories. Moreover, sector and stock categories are further filtered to represent the specific sectors and stocks categories. The finalized datasets are stored into two (.csv) files.
Steps to reproduce
- This dataset is prepared for news headlines related to Pakistan Stock Exchange (PSX) and contains news headlines along with publishing date. - Financial news headlines are scraped from daily "The News International" and properly categorized according to their area of influence in the stock market. -The news categorization scheme used for categorization is unsupervised. - News categorization scheme uses strength of NLP along with minimal manual effort. - Finalized dataset is validated using ANN based classification model.