Abstract
Abstract Because of exponential growth in the number of people who purchase products online, e-commerce organizations are vying for each other to offer innovative and improved services to its customers. Current platforms give its customers innovative services such as product recommendations based on their purchase histories and location, product comparison, and most importantly, a platform for expressing their experience and feedback. It is important for any e-commerce organization to analyze this feedback and to find out the sentiment of the customers for giving them better products and services. As large reviews may contain feedback in a mixed manner where a customer gives his opinion on different product features in the same review, finding out the exact sentiment is tedious. This work proposes aspect-specific sentiment analysis of product reviews using a well-sophisticated topic modeling algorithm called latent Dirichlet allocation (LDA). The topic words, thus, extracted are mapped with various aspects of an entity to perform the aspect-specific sentiment analysis on product reviews. Experiments with synthetic and real dataset show promising results compared to existing methods of sentiment analysis.
Highlights
Platforms such as social networks, micro-blogs, online reviews, and discussion forums are growing very fast, and the need for analyzing the sentiments of the users are increasing
Sentiment analysis is the process of extracting subjective information from natural language text, and it expresses the opinion or view of a user toward a topic
We show only the first five topics generated by the latent Dirichlet allocation (LDA), which are word unigrams
Summary
Platforms such as social networks, micro-blogs, online reviews, and discussion forums are growing very fast, and the need for analyzing the sentiments of the users are increasing. Sentiment analysis is the process of extracting subjective information from natural language text, and it expresses the opinion or view of a user toward a topic. LDA is the most popular and simplest topic model It is a generative for text and other collections of discrete data that generalizes or improves on several previous models including PLSI (probabilistic latent semantic indexing). LDA is a generative probabilistic model for collections of discrete data such as text corpora. It is a three-level hierarchical Bayesian model, in which each item of a collection is modeled as a finite mixture over an underlying set of topics. Wn), where wn is the nth word in the sequence, and a corpus is a collection of “M” documents denoted by D = W1 .
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
