The stock market is one of the most important economic participants. Stock market prediction has been an active area of research for a long time. There are many different ways to interpret the different stock market movements. Sentiments derived by users from news headlines have a tremendous effect on the buying and selling patterns of the traders as they easily get influenced by what they read. This analysis would be especially useful for buyers who are interested in short-term investment based on the financial news and listed companies which need a general idea of the daily market volatility.
This project started with web-scrapping stock price data from Yahoo Finance and stock news headlines from Financial Visualization for the companeis of interest. Data preprocessings, cleanings and visualizations were performed on the stock price data, while text preprocessing and sentiment analysis were conducted for the news healines data. Afterwards, the sentiment value was generated by using TextBlob and NLTK-Valder Lexicon tools. Classification models, such as logistic regression, SVM (Support-vector Machine), and random forest were trained and tuned on the data for predicting stock movements. Lastly, model accuracies were compared from different models using the test data. An insight and a discussion were constructed in the end.