Sentiment Analysis using a Machine Learning Approach in Python

Every day, people all over the world use social networking sites to express their thoughts on a broad range of issues. amongst the most well-known social networking sites is “Facebook” for exchanging messages with friends and family. People express their ideas, beliefs and experiences about circumstance. This research intends to create a prototype that analyzes people's attitudes towards the social difficulties of women, a topic that is becoming more relevant in many nations. Using a Facebook scraper written in Python, we were able to collect a dataset of Facebook data, which we subsequently cleaned up using the nltk toolkit. Machine learning techniques and approaches are used to study the feelings of individuals. It is possible in Python to classify each Facebook post as either good, negative, or neutral, depending on the polarity of its thoughts. The hashtags #Women and #MeToo were used to collect data. According to the results of the research, women's experiences, views, and concerns are more often expressed using the hashtag than those of males. The model was built and tested using a variety of machine learning techniques. The accuracy, recall, and f1-score of these models were evaluated using a variety of testing criteria. Furthermore, the performance of each model is compared to each other.