Information Technology Reference
In-Depth Information
similarity between them and the documents is computed as the sum of a tf/idf (term frequency/inverse
document frequency) score and user relevance feedbacks. Powerize Server (Kim, Oard & Romanik,
2000) is a content-based system which builds a model to take into account user's information needs.
This model is constructed explicitly by the user or implicitly inferring user behaviour. The proposed
system is based on parameters to define the user behaviour and starting from them and their correlations
the user model. In White, Ruthven & Jose (2002) a system for relevance feedback in Web retrieval is
presented. The authors follow two types of approaches based on explicit and implicit feedback. They
investigate on the degree of substitution between the two types of evidence. Using relevance feedback
the system displays new documents and in particular documents that have been retrieved but not yet
considered. The paper (Tan et al., 2007) studies the use of term-based feedback for interactive information
retrieval in the language modelling approach. The authors propose a cluster-based method for select-
ing terms to present to the user for judgment, as well as effective algorithms for constructing refined
query language models from user term feedback. The authors (Lad & Yang, 2007) propose news ways
of generalizing from relevance feedback through a pattern-based approach to adaptive filtering. The
patterns are wildcards that are anchored to their context which allows generalization beyond specific
words, while contextual restrictions limit the wildcard-matching to entities related to the user's query.
In (Chang & Chen, 2006) the authors present a method for query re-weighting to deal with document
retrieval. The proposed method uses genetic algorithms to re-weight a user's query vector, based on
the user's relevance feedback, to improve the performance of document retrieval systems. It encodes a
user's query vector into chromosomes and searches for the optimal weights of query terms for retrieving
documents by genetic algorithms. After the best chromosome is found, the proposed method decodes the
chromosome into the user's query vector for dealing with document retrieval. The proposed query re-
weighting method can find the best weights of query terms in the user's query vector, based on the user's
relevance feedback. In Russ et al. (2007) a relevance feedback technique for association rules which is
based on a fuzzy notion of relevance is proposed. The used approach transforms association rules into
a vector-based representation using some inspiration from document vectors in information retrieval.
These vectors are used as the basis for a relevance feedback approach which builds a knowledge base of
rules previously rated as (un)interesting by a user. Given an association rule the vector representation is
used to obtain a fuzzy score of how much this rule contradicts a rule in the knowledge base. This yields
a set of relevance scores for each assessed rule which still need to be aggregated. Rather than relying on
a certain aggregation measure the authors utilize OWA operators for score aggregation to gain a high
degree of flexibility and understandability.
Relevance feedback techniques are also used in other contexts, such as multimedia retrieval; e.g.,
in Zhang, Chai & Jin (2005) a text-based image retrieval system is described, Djordjevic & Izquierdo
(2007) introduce a framework for object-based semi-automatic indexing of natural images and Grigo-
rova et al. (2007) proposes an adaptive retrieval approach based on the concept of relevance-feedback,
which establishes a link between high-level concepts and low-level features.
ontologies
In past years, the ontological aspects of information have acquired a strategic value. These aspects
are intrinsically independent from information codification, so the information itself may be isolated,
recovered, organized, and integrated with respect to its content.
Search WWH ::
Custom Search