Interactive Deep Neural Network for Aspect-Level Sentiment Analysis

Abstract

Opinion mining, also known as sentiment analysis, is the act of analysing text written in natural language on a topic and categorizing it as positive, negative, or neutral based on the sentiments, emotions, and views stated in it. On the internet nowadays, the number of people who express their ideas through reviews is growing day by day. Manually analysing and extracting thoughts from such a large number of evaluations are nearly very complex and much tough to handle. The work analyses the deep learning techniques for extracting aspects in opinion mining in this research. Aspect extraction is a subtask of emotion investigation that needs discovering opinion targets in opinionated text, and identifying the precise characteristics of a product or service. A model for fine-grained aspect-based opinion mining is proposed that addresses important aspects of effective aspect-based opinion mining. Therefore, this work analyses and states the various methodologies that focuses on aspect based sentimental analysis. The implementation of this approach, is outside the scope of this paper.