Abstract
In the last decades the popularity of natural language interfaces to databases (NLIDBs) has increased, because in many cases information obtained from them is used for making important business decisions. Unfortunately, the complexity of their customization by database administrators make them difficult to use. In order for a NLIDB to obtain a high percentage of correctly translated queries, it is necessary that it is correctly customized for the database to be queried. In most cases the performance reported in NLIDB literature is the highest possible; i.e., the performance obtained when the interfaces were customized by the implementers. However, for end users it is more important the performance that the interface can yield when the NLIDB is customized by someone different from the implementers. Unfortunately, there exist very few articles that report NLIDB performance when the NLIDBs are not customized by the implementers. This article presents a semantically-enriched data dictionary (which permits solving many of the problems that occur when translating from natural language to SQL) and an experiment in which two groups of undergraduate students customized our NLIDB and English language frontend (ELF), considered one of the best available commercial NLIDBs. The experimental results show that, when customized by the first group, our NLIDB obtained a 44.69 % of correctly answered queries and ELF 11.83 % for the ATIS database, and when customized by the second group, our NLIDB attained 77.05 % and ELF 13.48 %. The performance attained by our NLIDB, when customized by ourselves was 90 %.
Highlights
Databases (DBs) are present everywhere, many applications access them at all times, and in many cases information obtained from them is used for making important business decisions
With a natural language interfaces to databases (NLIDBs), a user types a query in natural language, as he/ she would do when communicating with another person, and the interface interprets the query and translates it to a DB query language statement, which is submitted by the NLIDB to a DB management system to get the information requested
In this article we present results from an experiment aimed at comparing the effectiveness and easiness of customization of English language frontend (ELF) (Elf 2015) and our NLIDB, which is based on a new semantically-enriched data model described in Pazos et al (2011) and a layered architecture for the translation of NL queries to SQL
Summary
Databases (DBs) are present everywhere, many applications access them at all times, and in many cases information obtained from them is used for making important business decisions. The data dictionary (called semantic information dictionary) of our interface permits to carry out a customization, which consists of: relating words or phrases from different syntactic categories (such as nouns, verbs, adjectives and prepositions) to DB tables and columns, as well as defining imprecise values (those that represent a value range, such as morning, afternoon, evening) as equivalent to value ranges, and defining aliases (those values that can be denoted in different ways, such as: noon, equivalent to 12:00 hrs.; midnight, equivalent to 0:00 hrs.; dozen, equivalent to 12) as equivalent to specific values This functionality permits to solve problems not previously addressed by other NLIDBs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
