Stock often fluctuates based on the news headline. Briefly, evaluating the diversity of news headlines can provide valuable insights into the stock market's future. So, here sentiment of stock has been analyzed to predict a fall or rise in near future only using news paper headlines.
- Collected Dataset from Kaggle
- Preprocessed using NLTK
- Used CountVectorizer for word embedding
- Implemented Classifiers are - Random Forest, Multinomial Naive Bayes, Logistic Regression, K-Nearest Neighbors and Support vector Classifier
- Evaluating all the models, it is observed that Logistic Regression and Random Forest are outracing the rest of the models.
Finally, confusion matrices are generated to visualize and summarize the performance of all the classification algorithms.