A Novel Auto-Annotation Technique for Aspect Level Sentiment Analysis

In machine learning, sentiment analysis is a technique to find and analyze the sentiments hidden in text. For analysis, annotated data basic requirement. Generally, this manually annotated. Manual annotation time consuming, costly laborious process. To overcome these resource constraints research has proposed fully automated for aspect level analysis. Dataset created from reviews of ten most popular songs on YouTube. Reviews five aspects—voice, video, music, lyrics song, are extracted. An N-Gram based proposed. Complete dataset consists 369436 that took 173.53 s annotate using while might have taken approximately 2.07 million seconds (575 h) if it was manually. validation technique, sub-dataset—Voice, as well with technique. Cohen's Kappa statistics used evaluate degree agreement between two annotations. The high value (i.e., 0.9571%) shows two. This validates quality good manual even far less computational cost. also contributes consolidating guidelines