Stock market analysis and forecast is a very popular and profitable business area nowadays. In the world of stock exchanges one can notice a systematic fluctuation of prices. Hypothetically this system is influenced by the news related to different markets and the governmental sector.
With the spreading of the World Wide Web countless online news sources appeared providing information about the aforementioned topic. This type of so-called natural language information is considered to have very subjective content and therefore processing them is not as trivial as it looks.
In my thesis I collected and processed this kind of news and using text mining methods drawn the necessary conclusions. During the process I made a tone analysis on the different news articles and sorted them into three tone categories of positive, neutral and negative.
Comparing the processed news to the fluctuation of the stock exchange prices I’ve drawn a wide picture of link between these two elements.
The results show what kind of influence the online news can have over the stock prices of the biggest market participants.
To solve the problem I used Python programming language which’s support libraries provided me with the option of text mining, sentiment analysis and detection of different opinions.