An Approach to Sentimental Analysis of Drug Reviews using RNN-BiLSTM Model

In past years, large amount of data in various forms is available for use in medical domain to analyse various health conditions. The reviews in the medical industry are based on price, side- effects, user experience and proper dosage. The drug reviews are much more complicated as they contain medical terminologies like chemical used in drugs, medical health conditions etc. To overcome these issues this paper proposes a recommendation system based on a hybrid RNN stacked with bi-directional LSTM model and a gradient boosting framework, LightGBM classifier, which takes health condition as the input and recommends the best drugs based on the analysis of the reviews with 83% accuracy.

Keywords: Sentimental Analysis, Machine Learning, LightGBM, RNN, LSTM

Suggested Citation: Suggested Citation

Bemila, Theres and Kadam, Isha and Sidana, Anushka and Zemse, Shivani, An Approach to Sentimental Analysis of Drug Reviews using RNN-BiLSTM Model (April 8, 2020). Proceedings of the 3rd International Conference on Advances in Science & Technology (ICAST) 2020, Available at SSRN: https://ssrn.com/abstract=3599208 or http://dx.doi.org/10.2139/ssrn.3599208