Learning an optimal sequence of questions for the disambiguation of queries over structured data

Achim Rettingerf, Alexander Hagemann, Matthias Nickles

    Research output: Chapter in Book or Conference Publication/ProceedingConference Publicationpeer-review

    1 Citation (Scopus)

    Abstract

    Intelligent systems interacting with users often need to relate ambiguous natural language phrases lo formal entities which can be further processed. This work strives for learning an optimal sequence of disambiguation questions asked by an agent in order to achieve a perfect interactive disambiguation, setting itself off against previous work on interactive and adaptive dialogue systems for disambiguation in question answering. To this aim, we built a hybrid system that exhibits deductive and statistical inference capabilities by combining techniques from natural language processing, information retrieval, answer set programming and relational reinforcement learning.

    Original languageEnglish
    Title of host publicationMachine Learning for Interactive Systems
    Subtitle of host publicationBridging the Gap Between Perception, Action and Communication - Papers Presented at the 28th AAAI Conference on Artificial Intelligence, Technical Report
    PublisherAI Access Foundation
    Pages27-30
    Number of pages4
    ISBN (Electronic)9781577356684
    Publication statusPublished - 2014
    Event28th AAAI Conference on Artificial Intelligence, AAAI 2014 - Quebec City, Canada
    Duration: 28 Jul 2014 → …

    Publication series

    NameAAAI Workshop - Technical Report
    VolumeWS-14-07

    Conference

    Conference28th AAAI Conference on Artificial Intelligence, AAAI 2014
    Country/TerritoryCanada
    CityQuebec City
    Period28/07/14 → …

    Keywords

    • Disambiguation
    • Information retrieval
    • Linked Data
    • Named-Entity recognition
    • Question answering
    • Relational reinforcement learning

    Fingerprint

    Dive into the research topics of 'Learning an optimal sequence of questions for the disambiguation of queries over structured data'. Together they form a unique fingerprint.

    Cite this