I-SG : Interactive Search Grouping - Search result grouping using Independent Component Analysis

Thomas Lauritsen, Thomas Kolenda

AbstractWe present a computational simple and efficient approach to
unsupervised grouping the search result from any search engine. Along with each
group a set of keywords are found to annotate the contents. This
approach leads to an interactive search trough a hierarchial
structure that is build online. It is the users task to improve the search, trough expanding
the search query using the topic keywords representing the desired groups.
In doing so the search engine limits the space of possible search results, virtually moving down in the
search hierarchy, and so refines the search.
KeywordsInternet, WWW, Search, Datamine, Independent Component Analysis, ICA, classify, cluster, keywords
TypeTechnical report
Journal/Book/ConferenceFirst Annual Google Programming Contest
Year2002    Month April
PublisherDepartment of Mathematical Modeling, Technical University of Denmark
Electronic version(s)[pdf]
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing