Medication Recommendation System for Skin Diseases using NLP

Skin diseases are common health problems that affect a large number of people globally nowadays; due to busy schedules they are not able to take proper care. There is an imbalance among the healthcare experts and the patients, lack of time and trustworthy resources. Due to which individuals initiate self-administration of medicinal drugs without proper consultation, which makes the circumstance worse than normal. The recommended treatments for these diseases may vary depending on severity, type and cause of the condition. A proposed tool to tackle this problem is natural language processing (NLP). The aim of this paper is to examine the potential of NLP for prescribing medications for skin conditions. The proposed medication recommendation system analyses the user-reviews using Vader Sentiment Analysis to determine their sentiments and assigns a score to each review. The SVM algorithm with Gradient Descent and RNN with LSTM is used to classify the reviews into different categories based on their score and verifies the Vader score for recommendation of medications. Therefore, the aim is to provide accurate and personalised medications to patients suffering from skin diseases.