DBMSs Should Talk Back Too - Computer Science > DatabasesReportar como inadecuado




DBMSs Should Talk Back Too - Computer Science > Databases - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: Natural language user interfaces to database systems have been studied forseveral decades now. They have mainly focused on parsing and interpretingnatural language queries to generate them in a formal database language. Weenvision the reverse functionality, where the system would be able to take theinternal result of that translation, say in SQL form, translate it back intonatural language, and show it to the initiator of the query for verification.Likewise, information extraction has received considerable attention in thepast ten years or so, identifying structured information in free text so thatit may then be stored appropriately and queried. Validation of the recordsstored with a backward translation into text would again be very powerful.Verification and validation of query and data input of a database systemcorrespond to just one example of the many important applications that wouldbenefit greatly from having mature techniques for translating such databaseconstructs into free-flowing text. The problem appears to be deceivinglysimple, as there are no ambiguities or other complications in interpretinginternal database elements, so initially a straightforward translation appearsadequate. Reality teaches us quite the opposite, however, as the resulting textshould be expressive, i.e., accurate in capturing the underlying queries ordata, and effective, i.e., allowing fast and unique interpretation of them.Achieving both of these qualities is very difficult and raises severaltechnical challenges that need to be addressed. In this paper, we first exposethe reader to several situations and applications that need translation intonatural language, thereby, motivating the problem. We then outline, by example,the research problems that need to be solved, separately for data translationsand query translations.



Autor: Alkis Simitsis HP Labs, Yannis Ioannidis

Fuente: https://arxiv.org/







Documentos relacionados