Feature Based Opinion Mining

Suppose that a customer who has given a high rating about a mobile phone writes the following review about the product: "The front camera of the phone is excellent! Truly speaking, this is the best front camera I have experienced so far". From this review, we can understand two things. First, the customer holds a positive opinion about the phone. Secondly, the front camera of the phone is the targeted feature on which the opinions have been expressed in the review. The project particularly focussed in discovering the these patterns. A framework was designed which can enable one to discover the targets on which the opinions have been expressed in a review and then determine the polarity and strength of the opinions. This kind of detailed analysis can help one to realise the sentiments of the customers on various product features.

The Project Work

Prof. Gourab Nath first explored this problem with a team of four students - Navyashree ATM, Pankaj Nagpal, Astha Jaggi, Joyinee Dasgupta from the batch of July 2018. That time the problem was expored by using a pattern matching approach. This problem was later explored in details with a team from the July 2019 batch. The team comprized of five students - Ramya Ramprakash, Randeep Ghosh, Bhavya Sai Shree Vemavarapu, Manish Nair and their mentor Prof. Gourab Nath. The team showed an amazing demonstration on how this opinioin mining system may work backed up with necessary theoretical details. Their project got recognied as the "Best Project" of their batch.

Tech Talk at Cypher 2019

Prof. Gourab Nath further fine tuned this work and presented the work in a tech talk at Cypher 2019. Arpit Prakash, a student from the batch of Jan 2020, extended his helping hands to Prof. Nath do some further experimentations on this work.

Professor Gourab Nath at Cypher 2019

Here is a short description of the tech talk at Cypher 2019. The full video can be found here.

Paper Presentation in the proceedings of 7th International Conference of Business Analytics and Intelligence, IIM Bangalore

Prof. Gourab identified an application of this problem in the areas of customer segmentation and review search and extended this work to formulate a research problem. Randeep Ghosh, a member of the winning team, co-authored the work. This project has a third author, Rishav Nath, a student of MSc. Statistics from University of Kalyani. The paper got accepted and was presented in the International Confetence of Busines Analytics and Intelligence at IIM Bangalore on December 05, 2019. Click here to read the full paper.

Paper Presentation in the proceedings of 3rd National Conference of Machine Learning and Artificial Intelligence, IIM Bangalore

This work was further extended to explore opportunities to compre similar products based on user reviews. A system suggested by Prof. Bing Liu in 2006 was explored and analysed. A team of five students from the Jan 2020 batch - Arpit Prakash, Meenakshi Bhandari, Sushmita Umredkar, Vikas Yadav and Tejas Vartak under the mentorship of Prof Gourab Nath. The analysis brought out some interesting facts. The results was presented in the 3rd National Conference of Machine Learning and Artificial Intelligence organized by IIM Bangalore and was held at LBISM, Delhi.

Here is a video demonstration on how the prototype of this product will look like.