Improving intelligence through use of Natural Language Processing. A comparison between NLP interfaces and traditional visual GIS interfaces.

摘要

This paper aims at a comparative analysis between accessing GIS using traditional visual interfaces and new ones, which involve a Natural Language Processing (NLPj approach that provides support for the interaction between users and the information system. Lately, GIS have been used at a very large scale; also, users of these systems are different in many aspects, like technical skills, interests, goals, base of knowledge, etc. Among GIS features the location systems play a big role for ITS. To optimize urban mobility it has become more and more crucial to be able to offer location based services very accurate and available even for not expert use. In this scenario NLP allows to use un-precise expressions and queries, of synonyms and qualitative expressions in a very simple way, proposing itself as a powerful interface enabler.The metrics defined and considered by the authors to be relevant, from the points of view mentioned above, are: accessibility (technical skills required), speed (time required to identify the desired results) and quality (human perception of mistakes and the standardization of the formulas used to perform the queries). Experiments are performed using a traditional GIS interface (ArcGIS) and a NLP based interface called ASQL and provided by the company Ula. Ula's technology is being used for the project i-Tour which is developed under the 7th framework with the financial support of the CE.