I-Corps: An Interactive Query Interface

University Of Texas At Austin, Austin TX

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project will be to enable companies and end-users to more easily extract relevant information from data sources. This system can be beneficial for a wide range of companies including e-commerce, business intelligence, and database and could systems. It is envisioned that this technology will be used in two ways: First, companies can use the system internally to enable easier and faster data access for their employees, thereby reducing dependence on database administrators and cutting costs. Second, companies could use the system externally to provide a better search interface to their customers. In addition, this technology can potentially also be useful to academic researchers, particularly within social and natural sciences, by allowing them to easily extract relevant data. This I-Corps project provides an interface for extracting data from databases by writing English sentences. Although relational databases are the most common choice for the storage of information, retrieving a relevant subset of the data from them requires expertise in a formal query language. This technology allows end-users to query information stored in a relational database without having to learn a formal query language. The technology is based on a research which combines advanced natural language processing techniques with automated program synthesis and repair. A key advantage of this approach is that it is database-agnostic and does not require end-users to know the database schema. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.