• Publications
    • Conference Papers
    • Workshop Papers
    • Journal Papers
    • Publicity
    • Books
    • Theses
    • Submitted
  • Professional Activities
  • Teaching
  • About
  • Contact

Edgar Meij

semantic search research ッ

  • Publications
    • Conference Papers
    • Workshop Papers
    • Journal Papers
    • Publicity
    • Books
    • Theses
    • Submitted
  • Professional Activities
  • Teaching
  • About
  • Contact
TREC KBA logo

The University of Amsterdam at TREC 2012

09/11/2012 Blog Publications Unrefereed No Comments

This year the Information and Language Processing Systems (ILPS) group of the University of Amsterdam participated in the Microblog and the Knowledge Base Acceleration (KBA) tracks. In this paper, we describe our participation for each of these tracks, in two largely independent sections. We detail the runs we submitted, present the results of the submitted runs, and, where possible, provide an initial analysis of these results.

As to TREC KBA, this year’s single task concerned entity linking on a stream of data. That is, given a target entity from a knowledge base and an incoming stream consisting of textual content such as web pages, news items, and social media content, generate a score for each item based on how pertinent it is to the target entity. Recent advances have enabled a precise manner of analysis, where phrases occurring in documents are automatically linked to entries in a knowledge base. This process is commonly known as entity linking. Entity linking facilitates advanced forms of searching and browsing in various domains and contexts. It can be used, for instance, to anchor the textual resources in background knowledge; authors or readers of a piece of text may find entity links to supply useful pointers. Another application can be found in search engines, where it is increasingly common to link queries to entities and present entity-specific overviews. Typical solutions to entity linking operate on static collections of documents. In this year’s TREC participation, we adapt our entity linking system to be able to operate in this dynamic setting. To this end, we have implemented an incremental learning algorithm that updates at each time step, thereby improving performance at each point in time. Our algorithm also caters for different levels of aboutness.

  • [PDF] R. Berendsen, E. Meij, D. Odijk, M. de Rijke, and W. Weerkamp, “The University of Amsterdam at TREC 2012,” in Trec 2012 working notes, 2012.
    [Bibtex]
    @inproceedings{TREC:2012:wn,
    Author = {Berendsen, R. and Meij, E. and Odijk, D. and de Rijke, M. and Weerkamp, W.},
    Booktitle = {TREC 2012 Working Notes},
    Date-Added = {2012-10-28 22:15:47 +0000},
    Date-Modified = {2013-05-22 11:43:55 +0000},
    Month = {November},
    Series = {TREC 2012},
    Title = {{The University of Amsterdam at TREC} 2012},
    Year = {2012}}
context-based-entity-linking--university-of-amsterdam-at-tac-2012edgar_meijedgar-de-meijedgar-meijEntity linkingentity-linking-and-retrievalInformation retrievalLanguage modelingMachine learningmeij-2012-entitiesMicroblogspersonalized-time-aware-tweets-summarizationSemantic linkingText miningTREC KBATREC Microblogtrec-2012-twittertrec2012contextresultsTwitteryork-university-trec-2012

Getting ready for Horizon 2020 – Workshop notes

TREC 2012 summary

Leave a Reply Cancel reply

Time limit is exhausted. Please reload CAPTCHA.

Edgar Meij logo

Welcome!

This is the website of Edgar Meij. I lead several groups of researchers and engineers at Bloomberg working on knowledge graphs, question answering, information retrieval, machine learning, and more…

Search

Tweets by @edgarmeij

Tags

AIDA Artificial Intelligence CLEF DBpedia Document priors edgar-meij entity-linking-and-retrieval entity-linking-and-retrieval-tutorial entity-linking-tutorial Entity finding Entity linking Information retrieval Knowledge base population Knowledge Graph Language modeling Linking Open Data LOD logo-penerbit-buku-internasional Lucene Machine learning meij MeSH Microblogs penerbit-buku-internasional Query log analysis Query modeling Relevance modeling Semanticizing Semantic linking Semantic query analysis Semantic search Teaching Text mining TREC Blog TREC Enterprise TREC Genomics TREC KBA TREC Microblog TREC Relevance Feedback Tutorial Twitter Web services Wikipedia Workflows Workshop
Proudly powered by WordPress | Theme: Doo by ThemeVS.