Abstract
The performance and reliability of converting natural language into structured query language can be problematic in handling nuances that are prevalent in natural language. Relational databases are not designed to understand language nuance, therefore the question why we must handle nuance has to be asked. This paper is looking at an alternative solution for the conversion of a Natural Language Query into a Structured Query Language (SQL) capable of being used to search a relational database. The process uses the natural language concept, Part of Speech to identify words that can be used to identify database tables and table columns. The use of Open NLP based grammar files, as well as additional configuration files, assist in the translation from natural language to query language. Having identified which tables and which columns contain the pertinent data the next step is to create the SQL statement.
Highlights
With the quantity of real-time data and the speed of data increases the need to search and extract data from multiple sources is becoming more important
The keyword searching proposed in this paper unlike Jwalapuram & Mamidi [5] uses Part of Speech (POS) [5] processing and an index file which allows for individual words to be extracted from the natural language query
The solution proposed by this paper allows for the natural language query “What are the odds on a game involving caro?” to be converted into an Structured Query Language (SQL) statement using the following steps: Figure 4
Summary
College of Engineering, Design and Physical Sciences, Brunel University, London, UK. How to cite this paper: Skeggs, R. and Lauria, S. (2019) A Shallow Parsing Approach to Natural Language Queries of a Database.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have