C. Monz and M. de Rijke. The University of Amsterdam at CLEF 2001. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Creating a Dutch testbed to evaluate the - Retrieval From Textual (Correct)
....on letter 6 grams, all systems used some language specific software designed to handle Dutch texts. Figure 3 summarises the best recall precision averages of the participating systems. A brief description of the experiments is given below. AmsNlM The University of Amsterdam (Monz and De Rijke [14]) used their FlexIR system. The focus of their experiments was on the effects of morphological analyses such as stemming and compound splitting on retrieval effectiveness. They found as much as 55 improvement in average precision. Similar, but less spectacular, results were found if title only ....
C. Monz and M. de Rijke. The University of Amsterdam at CLEF-2001. In Peters [16], pages 165--170.
Source Code Retrieval using Conceptual Graphs - Mishne (2003) (1 citation) Self-citation (De rijke) (Correct)
No context found.
C. Monz and M. de Rijke. The University of Amsterdam at CLEF 2001.
Language-dependent and Language-independent.. - Kamps, Monz, de.. (2003) Self-citation (Kamps Monz De rijke Sigurbjornsson) (Correct)
No context found.
Kamps, J., Monz, C., de Rijke, M., Sigurbjornsson, B.: The University of Amsterdam at CLEF-2003. In Peters, C., ed.: Results of the CLEF 2003 Cross-Language System Evaluation Campaign. (2003) 71--78
Document Retrieval in the Context of Question Answering - Monz (2003) (1 citation) Self-citation (Monz) (Correct)
....55.0 68.5 7.0 5.0 3.0 mad 89.2 62.8 60.1 8.9 6.6 3.0 Table 1. The median number of relevant documents and the corresponding median absolute deviation (mad) 2. 2 Document Retrieval Approaches All retrieval techniques discussed in the remainder of this article use the FlexIR retrieval system [7]. FlexIR is a vector space retrieval system with several features including positional indexing, blind feedback, and structured querying. In this subsection we introduce some techniques which are known to have a positive impact on the effectiveness of document retrieval, and which have also been ....
C. Monz and M. de Rijke. The University of Amsterdam at CLEF 2001. In Working Notes for the Cross Language Evaluation Forum Workshop (CLEF 2001), pages 165--169, 2001.
The University of Amsterdam at INEX-2002 - Marx, Kamps, de Rijke (2002) Self-citation (Kamps De rijke) (Correct)
....and structure topics are discussed and an alternative is proposed. 1. INTRODUCTION The aim of our o#cial runs was to experiment with the effectiveness of di#erent types of morphological normalization for structured corpora. Morphological normalization proved successful for plain text collections [5, 6]. The XML retrieval task departs from the strict Boolean query matching used in traditional database theory, allowing for various gradations of relevance. In particular, related words like morphological variants should share some of their relevance. In order to study the precise e#ect of ....
C. Monz, J. Kamps, and M. de Rijke. The University of Amsterdam at CLEF-2002.
Computing with Meaning - de Rijke Self-citation (De rijke) (Correct)
....retrieval, as well as domain specific system evaluation. The Derive project [17] is one of the spin o#s of the Logic in Action project. Monz and de Rijke, the principal investigators in Derive, took part in the CLEF 2001 evaluation in three monolingual tasks: Dutch, German, and Italian [47]. We were particularly interested in the e#ects of shallow morphological analyses: stemming or lemmatization, and compound splitting. For the purposes of this chapter, we restrict ourselves to a brief description of our results for Dutch and German. All submitted runs used FlexIR, an information ....
....of shallow morphological analyses: stemming or lemmatization, and compound splitting. For the purposes of this chapter, we restrict ourselves to a brief description of our results for Dutch and German. All submitted runs used FlexIR, an information retrieval system developed by Christof Monz [47]. The main goal underlying FlexIR s design is to facilitate flexible experimentation with a variety of retrieval components and techniques. FlexIR is implemented in Perl; it is built around the standard UNIX pipeline architecture, and supports many types of scoring, indexing, and retrieval tools. ....
C. Monz and M. de Rijke. The University of Amsterdam at CLEF-
Tequesta: The University of Amsterdam's Textual Question.. - Monz, de Rijke (2001) (1 citation) Self-citation (Monz De rijke) (Correct)
....most current QA systems, Tequesta is built on top of a retrieval system. The first step is to build an index for the document collection, in this case the TREC 10 collection. Then the question is translated into a retrieval query which is posed to the retrieval system. For retrieval we use FlexIR [13], a vectorspace based retrieval system, described in Section 2.1. The retrieval system is used to identify a set of documents that are likely to contain the answer to a question posed to the system. The top 100 documents returned by FlexIR are processed by a partial parser described in Section ....
C. Monz and M. de Rijke. University of Amsterdam at CLEF
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC