Text Analysis Conference Knowledge Base Population (TAC KBP)

The Text Analysis Conference (TAC) is a series of evaluations and workshops organized to promote research in Natural Language Processing and related applications, by providing a large test collection, common evaluation procedures, and a forum for organizations to share their results. The goal of TAC KBP is to develop and evaluate technologies for building and populating knowledge bases (KBs) about named entities from unstructured text. KBP systems must either populate an existing reference KB, or else build a KB from scratch.

Since 2009, LDC has supported TAC KBP through the development of source data and annotations, and by contributing to task development activities and discussions each year. In 2013, LDC is supporting six KBP tracks, all aimed at improving the ability to automatically populate knowledge bases from text:

    Entity Linking

    The entity linking task is to link names in a document collection to entities in a reference KB, or to new named entities discovered in the document collection. Tasks are offered in English, Spanish, and Chinese.

    English Slot Filling

    The slot filling task is to search a document collection to fill in values for predefined slots (attributes) for a given entity in a reference KB.

    Temporal Slot Filling

    The goal of temporal slot filling is to augment a KB with temporal constraints on slot filling relations.

    Cross-lingual Spanish Slot Filling

    The Spanish slot filling track extends English slot filling to the cross-lingual paradigm, in which English and Spanish documents are used to populate an English reference KB.

    Sentiment Slot Filling

    The overall goal of the Sentiment Slot Filling track is to assess the quality of detectors for scoped and attributed sentiment.

    Cold Start KBP

    The Cold Start track integrates entity linking and slot filling to build a knowledge base from scratch.

    Additional Information

    Xuansong Li, Stephanie M. Strassel, Heng Ji, Kira Griffitt, Joe Ellis
    Linguistic Resources for Entity Linking Evaluation: from Monolingual to Cross-lingual
    LREC 2012: 8th International Conference on Language Resources and Evaluation, Istanbul, May 21-27
     Available: Paper in PDF

    Paul McNamee, Hoa Trang Dang, Heather Simpson, Patrick Schone and Stephanie M. Strassel
    An Evaluation of Technologies for Knowledge Base Population
    LREC 2010, In Proceedings of the Seventh conference on International Language Resources and Evaluation. Valletta, Malta, May 2010
    Available: Paper in PDF

    Heather Simpson, Stephanie Strassel, Robert Parker, Paul McNamee
    Wikipedia and the Web of Confusable Entities: Experience from Entity Linking Query Creation for TAC 2009 Knowledge Base Population
    LREC 2010, In Proceedings of the Seventh conference on International Language Resources and Evaluation. Valletta, Malta, May 2010
    Available: Slides in PDF