Luís Miguel Cabral

Learn More
Use of the Web brought 32% (8/25) increase in performance Experiments Coping with five kinds of anaphoric expressions (see paper) Invoking PALAVRAS, a broad-coverage parser for Portuguese [Bick 2000] Dealing with sentence arguments (object, subject, etc.) as candidates Shallow reference resolution algorithm: 71% accuracy in general (over 122 questions)(More)
In this paper we provide a full overview of GikiCLEF, an evaluation contest (track) that was specifically designed to expose and investigate cultural and linguistic issues involved in multimedia collections and searching. In GikiCLEF, 50 topics were developed by a multilingual team with non-English users in mind. Answers should be found in Wikipedia, but(More)
In this paper we describe GikiCLEF, the first evaluation contest that, to our knowledge, was specifically designed to expose and investigate cultural and linguistic issues involved in multimedia collections and searching, and which was organized under the scope of CLEF 2009. We present the task, its motivation, the results of the participants and the GIRA(More)
This overview paper is devoted to a critical assessment of GikiCLEF 2009, an evaluation contest specifically designed to expose and investigate cultural and linguistic issues in Wikipedia search, with eight participant systems and 17 runs. After providing a maximally short but self contained overview of the GikiCLEF task and participation, we present the(More)
SUPeRB is a digital library helper that aims at updating and maintaining specific publication repositories, and assisting in the publishing of publication records, for institutions and individual actors. It gathers bibliographic data from Web pages and documents and integrates that data into a local repository of bibliographic data on a specific domain. By(More)
Esfinge is a general domain Portuguese question answering system that participated in the last four editions of CLEF. This system uses the Web as a fundamental resource in its architecture, using information redundancy rather than sophisticated annotations of the document collections to retrieve answers. In this paper we describe experiments that took as(More)
  • 1