unsupervised-stock-market-tweets-sentiment-analysis

scrapping and labeling of stock market tweets using word embedings(word2vec) and unsupervised learning(K-means)

to do

• automate data pipeline for scraping, cleaning and modeling data in real-time: airflow, kafka
• create dashbord and include insights of financial data : dash, yahoo finance
• create emebdings using doct2vec for comparison with word2vec : gensim
• fine tune bert ,for sentiment analysis, on the newly labeled data : huggingface transformers