With the increased popularity of zero-commission investing and trading apps such as Robinhood and rise of retail traders, the influence of social media on financial markets have grown. Twitter was one of the most used social media platforms for this purpose. First, we used a machine learning classifier to predict the popularity of a tweet. Following that, a scenario was created where the stocks of S&P 500 companies that were mentioned in the popular tweets were bought and held over a certain period of time. The returns from each stock purchase were then used to train a machine learning model for obtaining recommendations to buy. Traders could potentially use the stock recommendations from our model to plan out investment strategies and build stock portfolios.