popular this month

recent comments

recent articles

  • The Avengers

    Almer S. Tigelaar 11 / 05 / 2012

    Marvel teased us with the release of this film near the end of various previously released super hero flicks like Captain America and Iron Man 2. This would be the movie that unites all the super heroes from the Marvel universe. Well actually, only those that had not been previously licensed to other studios. Hence, you will not find characters from X-Men, Spiderman, or the Fantastic Four in this movie. Director Joss Whedon brings back fond memories of creative television series like Firefly and Dollhouse, but what does he make of a 220 million blockbuster production?

    read more 0 comments

  • Hugo

    Almer S. Tigelaar 06 / 03 / 2012

    Hugo is based on a relatively recently released (2007) award winning book by Brian Selznick. It is not surprising that the film rights to the books were quickly sold, and certainly not by the least of directors either: Martin Scorsese. He has a career spanning decades and has directed a string of movies in recent years which I liked, among which are Shutter Island, The Departed and Gangs of New York. However, those were admittedly all in different, less family friendly, genres. So, I went to Hugo hoping to be pleasantly surprised.

    read more 0 comments

  • How long would it take to read Wikipedia?

    Almer S. Tigelaar 21 / 02 / 2012

    Wikipedia has become the de facto encyclopedia on the Internet. A traditional encyclopedia spans many textbook volumes which would take any normal person ages to read. Few people would likely engage in such an endeavor. However, since Wikipedia is readily accessible: should you take up the challenge?

    read more 0 comments

Almer S. Tigelaar » Graduation Committees

Bertold van Voorst: Cluster-based Collection Selection in Uncooperative Distributed Information Retrieval

Almer S. Tigelaar 13 / 07 / 2010, 14:00

Cluster-based Collection Selection in Uncooperative Distributed Information Retrieval
by Bertold van Voorst

View in Repository

Abstract
The focus of this research is collection selection for distributed information retrieval. The collection descriptions that are necessary for selecting the most relevant collections are often created from information gathered by random sampling. Collection selection based on an incomplete index constructed by using random sampling instead of a full index leads to inferior results.

In this research we propose to use collection clustering to  compensate for the incompleteness of the indexes. When collection clustering is used we do not only select the collections that are considered relevant based on their collection descriptions, but also collections that have similar content in their indexes. Most existing cluster algorithms require the specification of the number of clusters prior to execution. We describe a new clustering  algorithm that allows us to specify the sizes of the produced clusters instead of the number of clusters.

Our experiments show that that collection clustering can indeed improve the performance of distributed information retrieval systems that use random sampling. There is not much difference in retrieval performance between our clustering algorithm and the well-known k-means algorithm. We suggest to use the algorithm we proposed because it is more scalable.

More in Graduation Committees: