Natural Language Query to NoSQL Generation Using Query-Response Model
Suravi Mondal, Prasenjit K. Mukherjee, Baisakhi Chakraborty, Rezaul Bashar · 2019
This paper proposes an automated query-response model termed Natural Language Query to NoSQL Generation using Query-Response Model (NLNSM) that can manage various types of natural language queries from the user. The NLNSM System uses POS Tagging and combination based technique to handle assertive, interrogative, imperative, compound and complex type query sentences from the user. The algorithm of NLNSM generates NoSQL query for each of the natural language queries (NL) to retrieve data from the MongoDB database resident within the NLNSM. The NLNSM system is also able to extract exact noun or noun phrases from natural language queries posted by the user without any interruption. The usage of MongoDB has added advantage for the system. Unlike traditional databases, MongoDB uses JSON Scripts or Binary JSON (also termed as BSON) which is smaller and faster for computation with easier parsing techniques. Moreover, in MongoDB, one collection holds different documents where the number of fields, content and size of the document can differ from one document to another. It also supports dynamic queries on documents, is highly scalable and enables faster and efficient access of datasets.