
Usually, Users are not aware of all the details about the hospitals, doctors, treatment or symptoms regarding the particular disease. For the small problems, users have to go personally to the hospital for a check-up which is time consuming and expensive. Also handling the telephonic calls for the complaints is quite hectic. Such problems can be solved by using medical Chabot which provides proper guidance for healthy living. The proposed medical chatbot functioning based on Natural language processing, topic modeling, aspect mapping and SVM which helps users to submit their queries about health problems, concerned specialists and get the suggestion about treatment and related services. The User can ask any personal query related to health care through medical chatbot without accessing the hospital in person. The proposed approach for medical chatbot has three phase processing. The first phase is the preprocessing stage which includes Natural language processing methods like word splitting, filter out punctuations, stop word removal and finally porter stemming is done to identify the root word. The second phase includes topic modeling, Aspect extraction and topic aspect mapping. Here the identified topics and aspects are mapped together and assigned to each categorized dataset and trained a machine with SVM. The third phase of the proposed approach is the trained system identifies the aspect from the human typed sentences in medical chatbot and these identified aspects matched with the database sets. And the chatbot redirected to matched dataset and retrieved the corresponding information and displayed in chatbot. Finally, check the efficiency of proposed work with different size of the medical conversation dataset and also the efficiency of the machine learning approach.