Stock Market Prediction Using Microblogging Sentiment Analysis and Machine Learning

The use of Machine Learning (ML) and Sentiment Analysis (SA) on data from microblogging sites has become a popular method for stock market prediction. In this work, we developed model predicting movement utilizing SA Twitter StockTwits data. Stock sentiment were used to evaluate approach validate it Microsoft stock. We gathered tweets StockTwits, as well financial Finance Yahoo. was applied tweets, seven ML classification models implemented: K-Nearest Neighbors (KNN), Support Vector (SVM), Logistic Regression (LR), Naïve Bayes (NB), Decision Tree (DT), Random Forest (RF) Multilayer Perceptron (MLP). main novelty work is that integrates multiple methods, emphasizing the retrieval extra features social media (i.e., public sentiment), improving prediction accuracy. best results obtained when analyzed using Valence Aware Dictionary sEntiment Reasoner (VADER) SVM. top F-score 76.3%, while Area Under Curve (AUC) value 67%.