VOICE CONTROLLED PERSONAL HEALTH ASSISTANCE DEVICE USING NATURAL LANGUAGE PROCESSING AND RANDOM FOREST DISEASE CLASSIFIER

Abstract: Health assistant devices are a need of the hour. In this paper, we outline one such device that has been developed. It is voice-controlled device that uses Google‟s speech to text converter API and Natural Language Processing for the communication with the user. The user provided input symptoms are classified into a disease using a Random Forest classifier. The NLP and classification are performed within the Raspberry Pi which acts as a standalone device. Temperature sensor is also interfaced for better user experience. The developed classifier has the accuracy of 95.02% which has further scope of improvement with a larger data set.