A multithreaded conversational interface for pedestrian navigation and question answering

Srinivasan Janarthanam, Oliver Lemon, Xingkun Liu, Phil Bartie, William Mackaness, Tiphaine Dalmas

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)
11 Downloads (Pure)

Abstract

We demonstrate a conversational interface that assists pedestrian users in navigating within urban environments and acquiring tourist information by combining spoken dialogue system, question-answering (QA), and geographic information system (GIS) technologies. In contrast to existing mobile applications which treat these problems independently, our Android agent addresses the problem of navigation and touristic question-answering in an integrated fashion using a shared dialogue context with multiple interleaved dialogue threads. In this paper, we present the architecture and features of our latest system, extended from an earlier version which was built and evaluated with real users (Janarthanam et al., 2013). The new features include navigation based on visible landmarks, navigation adapted to the user's previous route knowledge, and tourist information pushing based on visible and proximal points-of-interest. The system also uses social media to infer "popularity" of geographical entities.

Original languageEnglish
Title of host publicationProceedings of the 14th Annual Meeting of the Special Interest Group on Discourse and Dialogue
PublisherAssociation for Computational Linguistics
Pages151-153
Number of pages3
ISBN (Electronic)9781937284954
Publication statusPublished - 2013
Event14th Annual Meeting of the Special Interest Group on Discourse and Dialogue 2013 - Metz, France
Duration: 22 Aug 201324 Aug 2013

Conference

Conference14th Annual Meeting of the Special Interest Group on Discourse and Dialogue 2013
Abbreviated titleSIGDIAL 2013
CountryFrance
CityMetz
Period22/08/1324/08/13

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Modelling and Simulation
  • Human-Computer Interaction

Fingerprint Dive into the research topics of 'A multithreaded conversational interface for pedestrian navigation and question answering'. Together they form a unique fingerprint.

  • Cite this

    Janarthanam, S., Lemon, O., Liu, X., Bartie, P., Mackaness, W., & Dalmas, T. (2013). A multithreaded conversational interface for pedestrian navigation and question answering. In Proceedings of the 14th Annual Meeting of the Special Interest Group on Discourse and Dialogue (pp. 151-153). Association for Computational Linguistics.