ABSTRACT
OBJECTIVES: Local hospitals are operated by the resigned association of patients as passive communication channels. The online hospital data related to the users' queries are not transparent and reliable. Therefore, it is crucial to have an intelligent web chatbot that manages user requests and provides quick access to local hospital information. In this paper, we present a framework and functionality of a chatbot developed using web technologies. METHODS: The bot engine was integrated by several machine learning approaches like gradient descent (GD) and natural language processing (NLP) algorithms. The trained data entered into the bot were split into mini-word batches, and the GD algorithm was applied sequentially on each mini-batch. The NLP methods involved in converting a word to its stem with a text result less readable by humans. RESULTS: The employed ML algorithms were successfully incorporated to manage the alternative synchronisation of text and voice messages. CONCLUSIONS: The proposed bot can be a better solution for data extraction from local hospital which functioning as a good communication channel for both users and hospital staff and helpful in reducing the crowd.
