Pairwise similarity of TopSig document signatures
This paper analyses the pairwise distances of signatures produced by the TopSig retrieval model on two document collections. The distribution of the distances are compared to purely random signatures. It explains why TopSig is only competitive with state of the art retrieval models at early precision. Only the local neighbourhood of the signatures is interpretable. We suggest this is a common property of vector space models.
Impact and interest:
Citation counts are sourced monthly from and citation databases.
Citations counts from theindexing service can be viewed at the linked Google Scholar™ search.
|Item Type:||Conference Paper|
|Keywords:||Signature Files, Search Engines, Document Clustering, Topology, Vector Space IR, Random Indexing, Document Signatures, Near Duplicate Detection, Relevance Feedback, Random Projection|
|Subjects:||Australian and New Zealand Standard Research Classification > INFORMATION AND COMPUTING SCIENCES (080000) > ARTIFICIAL INTELLIGENCE AND IMAGE PROCESSING (080100) > Pattern Recognition and Data Mining (080109)
Australian and New Zealand Standard Research Classification > INFORMATION AND COMPUTING SCIENCES (080000) > LIBRARY AND INFORMATION STUDIES (080700) > Information Retrieval and Web Search (080704)
|Divisions:||Current > Schools > School of Electrical Engineering & Computer Science
Current > QUT Faculties and Divisions > Science & Engineering Faculty
|Copyright Owner:||Copyright 2012 [please consult the authors]|
|Deposited On:||11 Nov 2012 22:59|
|Last Modified:||20 Feb 2013 05:30|
Repository Staff Only: item control page