Classifying YouTube Comments Based on Sentiment and Type of Sentence

“However, not all YouTubers have enough time to go through all the comments on individual video. On the contrary, they must read all the comments to fully understand the public interest on their content. The solution to this inconvenience is addressed in our work. Our approach is to extract all the comments from a video and categorize them into multiple categories based on both sentiment and sentence types: Negative, Positive, Interrogative, Imperative, Corrective, and Miscellaneous. These categories can help YouTubers focus only on those comments that suit their interest.

Neural networks [21] can be used as a potential solution, however, they are difficult to tune and are not readily explainable. Explainability is especially important for comments that fall under multiple categories to clearly understand why a resulting category was selected.”