Order‐theoretical ranking

Abstract

Current best-match ranking (BMR) systems perform well but cannot handle word mismatch between a query and a document. The best known alternative ranking method, hierarchical clustering-based ranking (HCR), seems to be more robust than BMR with respect to this problem, but it is hampered by theoretical and practical limitations. We present an approach to document ranking that explicitly addresses the word mismatch problem by exploiting interdocument similarity information in a novel way. Document ranking is seen as a query-document transformation driven by a conceptual representation of the whole document collection, into which the query is merged. Our approach is based on the theory of concept (or Galois) lattices, which, we argue, provides a powerful, well-founded, and computationally-tractable framework to model the space in which documents and query are represented and to compute such a transformation. We compared information retrieval using concept lattice-based ranking (CLR) to BMR and HCR. The results showed that HCR was outperformed by CLR as well as by BMR, and suggested that, of the two best methods, BMR achieved better performance than CLR on the whole document set while CLR compared more favorably when only the first retrieved documents were used for evaluation. We also evaluated the three methods' specific ability to rank documents that did not match the query, in which case the superiority of CLR over BMR and HCR (and that of HCR over BMR) was apparent.

A ranking algorithm for query expansion based on the term's appearing probability in the single document

“…Latent semantic indexing, originally used for the dimension reduction of a term vector (Deerwester et al, 1990), is aimed at discovering the correlations among terms. Formal concept analysis (Carpineto and Romano, 2000), Phrasefinder (Jing and Croft, 1994), and similarity thesauri (Qiu and Frei, 1993) analyse and apply synonyms, hypernyms and hyponyms of terms. The whole document set must be examined and analysed in order to use global analysis techniques.…”

Research in concept lattice based automatic document ranking

“…Many researchers are using this well-founded formalism as a basis for measuring query-document relevance in text retrieval, i.e. Concept lattice-based ranking (CLR) [6,7] . The advantages of this approach lies as follows.…”

“…FCA generates a conceptual hierarchy of the domain by finding all possible formal concepts that reflect the relationships between attributes and objects [13] . [6,7] As input data, we consider a binary relation between a set of documents (D) and a set of terms (T), usually called context in the concept lattice theory. In the information retrieval field, this is the usual document by term relation.…”

“…Definition1, Definition2, Definition3 and Definition4 can be found in Carpineto, C. and Romano, G's papers [6,7] . We just introduce the following definitions to illustrate the traditional concept lattice based ranking (CLR).…”

A Digital Assets Navigating System Based on FCA

“…Now we can see several running FCA-based IR prototype (e.g., [2], [3], [4], [5] [6] ).Over the last few years, the range of FCA-based application has been expanded to new tasks such as automatic text ranking and IR from semi-structured data (e.g., [7], [8] ); at the same time, new IR domains have been investigated including email messages, web documents, and file systems [9] .…”