Democratic data fusion for information retrieval mediators

@article{Tzitzikas2001DemocraticDF,
  title={Democratic data fusion for information retrieval mediators},
  author={Yannis Tzitzikas},
  journal={Proceedings ACS/IEEE International Conference on Computer Systems and Applications},
  year={2001},
  pages={530-536}
}
  • Yannis Tzitzikas
  • Published 25 June 2001
  • Computer Science
  • Proceedings ACS/IEEE International Conference on Computer Systems and Applications
Abstract: Our research presented in this paper concerns the problem of fusing the results returned by the underlying systems to a mediating retrieval system, also called meta-retrieval system, meta-search engine, or mediator. We propose a fusion technique which is based solely on the actual results returned by each system for each query. The final (fused)ordering of documents is derived by aggregating the orderings of each system in a democratic manner. In addition, the fused ordering s… 

Figures from this paper

Query Translation for Mediators over Ontology-Based Information Sources
TLDR
This work proposes a model for providing integrated and unified access to multiple information sources and focuses on query translation between the mediator and the sources.
Collaborative ontology-based information indexing and retrieval
TLDR
The aim of this work is to propose formally sound methods for building such an information network based on ontologies which are widely used and are easy to grasp by ordinary Web users.
Mediators over ontology-based information sources
TLDR
The querying and answering process in a model for providing integrated and unified access to multiple information sources of the kind of Web catalogs is studied and algorithms for handling the main tasks of the mediator are presented.
Revisiting Optimal Rank Aggregation: A Dynamic Programming Approach
TLDR
This paper proposes an algorithm based on dynamic programming which, using relevancy information, obtains the aggregated list with the maximum performance that could be possibly achieved by any aggregation strategy.
Query Evaluation for Mediators over Web Catalogs
TLDR
This work presents a model for building mediators over Web catalogs, so as to provide users with cus-tomized views of such catalogs and focuses on query evaluation and complexity of query answering by the mediator.
On Ranking RDF Schema Elements (and its Application in Visualization)
TLDR
Several metrics for evaluating automatic methods for ranking schema elements are proposed and discussed, and the creation of a test collection for evaluating such methods is described, upon which several ranking methods for RDF schemas are evaluated.
Comparison and fusion of retrieval schemes based on different structures, similarity measures and weighting schemes
TLDR
The overall results show that the best thesis structure is title and the best similarity measure is Cosine with LTU weighting scheme, which is also the best combination for similarity measures combinations.
Image retrieval using automatic region tagging
TLDR
This work addresses the challenge of tagging, annotating or labelling image content automatically with semantic keywords based on the objects that they contain and the role that semantic keywords play in image retrieval.
Complete Proceedings
TLDR
This workshop aims to assess the state-of-the-art of Blended-Learning in SMEs and to identify research gaps and opportunities, in the form of best-practice reports, guidelines and heuristics.
Indexation et fusion multimodale pour la recherche d'informations par le contenu : Application aux bases de données d'images médicales
Dans cette these, nous nous interessons aux methodes de recherche d'information dans des bases de donnees constituees de documents multimedia. Nous les appliquons a des documents contenant des images
...
...

References

SHOWING 1-10 OF 17 REFERENCES
The Collection Fusion Problem
TLDR
This paper examines two collection fusion techniques that use the results of past queries to compute the number of documents to retrieve from each of a set of subcollections such that the total number of retrieved documents is equal to N, the number to be returned to the user.
Evaluation of an inference network-based retrieval model
TLDR
Network representations show promise as mechanisms for inferring probable relationships between documents and queries and have been used in information retrieval since at least the early 1960s.
Merging Results From Isolated Search Engines
TLDR
Two new techniques for merging search results are introduced: Feature Distance ranking algorithms and Reference Statistics, which are found to be more eeective than existing methods in an isolated-server environment such as the World Wide Web.
A decision-theoretic approach to database selection in networked IR
TLDR
A general decision-theoretic model is developed and a formula which estimates the number of relevant documents in a database based on dictionary information is derived, which is similar to the single-broker case.
Multi-Engine Search and Comparison Using the MetaCrawler
TLDR
The MetaCrawler provides a single, central interface for Web document searching that facilitates customization, privacy, sophisticated ltering of references, and more and serves as a tool for comparison of diverse search services.
Adaptive Agents for Information Gathering from Multiple, Distributed Information Sources
TLDR
An intelligent, adaptive Web search tool that can not only locate relevant information sources for the user, but also adapt to the frequent changes of the dynamic Web environment is described.
Multi-Service Search and Comparison Using the MetaCrawler
TLDR
The MetaCrawler provides a single, central interface for Web document searching that facilitates customization, privacy, sophisticated ltering of references, and more and serves as a tool for comparison of diverse search services.
Relevance weighting of search terms
TLDR
This paper examines statistical techniques for exploiting relevance information to weight search terms using information about the distribution of index terms in documents in general and shows that specific weighted search methods are implied by a general probabilistic theory of retrieval.
SAVVYSEARCH: A Metasearch Engine That Learns Which Search Engines to Query
TLDR
The SAVVYSEARCH metasearch engine is designed to efficiently query other search engines by carefully selecting those search engines likely to return useful results and responding to fluctuating load demands on the web.
A belief network model for IR
TLDR
A belief network model for IR which is derived from probabilistic considerations over a clearly defined sample space is introduced and how to extend the model with information from other queries to yield improved retrieval performance is shown.
...
...