Voice Based Database Analyzer

Abstract: In recent few years, Artificial Intelligence has shown significant improvement and its user cases are growing rapidly. An application area of Artificial Intelligence is Natural Language Understanding (NLU). Voice assistants can have a conversation with the users in natural language. Voice assistants are quite easy to use and can handle many difficult tasks easy. But currently there are no voice assistants or chatbot to analyze database data and also it hard to analyze database data for people with non-tech background.

Almost all Artificial Intelligence programs are reasoning programs. And, to the extent that they reason well about a problem, all are somewhat good at problem solving. Expert systems have a knowledge-base containing extracted experience and an inference engine which is essentially a set of rules for applying the knowledge base to each particular situation that is described to the program. The chatbot’s capabilities can be made better with modifications to the knowledge base or to the inference rules.

Traditionally chatbots were made on a decision tree-based logic. They contain many if else logics and these logic controls the flow of the tree-based logic. This tool is also based on the same logic where we try to extract user intent based on words present in the query.

In this project we are trying to make a voice-based database analyzer which will be an expert system. The expert system will have a knowledge base which will be used for database processing. It will take a database connection string as input then users will speak their queries, the queries will be converted from voice to text and it will perform the necessary calculations with the database data, knowledge base and queries and will show the text, tables or charts accordingly.