Spam Comments Detection on Instagram Using Machine Learning and Deep Learning Methods

The more popular a public figure on Instagram (IG), the number of followers also increase. When posts something, there are many comments from other users. In fact, all comments, not them relevant to post, such as advertising, links, or clickbait comments. type that irrelevant post is usually called spam Spam will interfere with information flow and may lead misleading information. This research compares machine learning (ML) deep (DL) classification methods based our collected Indonesian IG comment dataset. was conducted in following steps: dataset preparation, pre-processing, simple normalization, features generation using TF-IDF word embedding, application ML DL methods, performance evaluation, comparison. authors compare accuracy, F-1, precision, recall results. shows do significantly differ. Linear SVM, Extreme Tree (ET), Regression, Stochastics Gradient Descent algorithms can reach accuracy 0.93. At same time, method has highest 0.94 SimpleTransformer BERT architecture. difference between different.