Dept-of-Comp-Sci-University-of-Ghana/tweet-sentiment-extraction-cmsm_gp-12

tweet-sentiment-extraction-cmsm_gp-12 created by GitHub Classroom # Tweet Sentiment Extraction This project work with the dataset consiting of tweets that are either, positive, neutral or negative. The aim of this project is to predict the sentiment of tweets using different machine learning models. **Preprocessing** To obtain the best results possible, we chose to performe some preprocessing to our data before fitting the models. The chosen preprocessings we did was as follows: Setting the sentiment to numerical values; negative -1, neutral 0, positive 1 Removing stop words **Models** We chose the following models: Logistic Regression with TD-IDF Logistic Regression with Count Vectorizer ## Findings **Logistic Regression with TD-IDF** Accuracy Score: 69.998% **Logistic Regression with Count Vectorizer** Accuracy Score: 69.031% ## Challenges We had some challenges with displaying the results in an orderly fashion and creating a correct submission file. We did a quick fix, which might not meet the required format. At least the code should be right.