Journal article
Adaptive web search : Evolving a program that finds information
IEEE intelligent systems, Vol.21(5), pp.72-77
09/2006
DOI: 10.1109/MIS.2006.86
Abstract
An approach for automatically evolving new retrieval algorithms based on a user's evaluation of previously viewed documents by using genetic programming is discussed. The algorithm is based on genetics and uses selection and modification to breed increasingly effective retrieval algorithms. Genetic algorithms has been used because of their intrisically parallel search mechanisms and powerful global exploration capabilities in a high dimensional-space. It can help to generate a program that effectively combines document features for a particular user with a persistent information need. Thirty five separate mini-experiments, and a training data set was used to train the algorithm and validation to pick the best tree available from the training phase.
Details
- Title: Subtitle
- Adaptive web search : Evolving a program that finds information
- Creators
- Michael Gordon - University of Michigan–Ann ArborWEIGUO Fan - Virginia TechPraveen Pathak - University of Florida Health
- Resource Type
- Journal article
- Publication Details
- IEEE intelligent systems, Vol.21(5), pp.72-77
- Publisher
- IEEE Computer Society
- DOI
- 10.1109/MIS.2006.86
- ISSN
- 1541-1672
- eISSN
- 1941-1294
- Language
- English
- Date published
- 09/2006
- Academic Unit
- Business Analytics
- Record Identifier
- 9984380492702771
Metrics
1 Record Views