A Novel Solution for the Aggregation Problem in Natural Language Interface to Databases (NLIDB)

Abstract

Natural Language Interface to Databases (NLIDB) systems usually do not deal with aggregations, which can be of two types: aggregation functions (such as count, sum, average, minimum, and maximum) and grouping functions (GROUP BY). This paper addresses the creation of a generic module, to be used in NLIDB systems, that allows such systems to perform queries with aggregations, on the condition that the query results the NLIDB returns are or can be transformed into tables. The paper covers aggregations with specificities, such as ambiguities, timescale differences, aggregations in multiple attributes, the use of superlative adjectives, basic unit measure recognition, and aggregations in attributes with compound names.

Query Answer Reformulation over Knowledge Bases

“…Finally, several studies addressed the problem of creating a Natural Language interface to databases, such as [Novello and Casanova 2020]. Usually, the proposed process has four steps: Question Analysis, Phrase Mapping, Disambiguation, and Query Construction -not necessarily in this order [Diefenbach et al 2018].…”