Automating Web Scraping of User Comments for Sentiment Analysis in Social Networks

Social networks can show the general mood of the population or groups interested in a specific topic. Machine learning will recognize people's moods in posts and automatically classify them according to preferences. Often, the problem is a lack of training data, which can now be solved using ChatGPT. Before use, the training data is preprocessed and verticalized. Web scraping from real social network data can be used as test data to evaluate neural network performance. A decision tree machine learning algorithm is used for data classification. The study's results may be of interest to analysts and page managers on social networks Facebook and Instagram.