Abstract
The selection of information sources in a distributed information retrieval environment remains a critical issue. In this context, it is known that a distributed information retrieval system consists of a huge number of sources. Ensuring retrieval effectiveness is to search only sources which are likely to contain relevant information for a query. An important number of heuristics exist among which we quote genetic algorithm that is used to solve the above problem. The proposed genetic algorithm consists in finding the best selection in large space of potential solutions; where a solution is represented as a combination of a set of sources. The improvement of selection accuracy is assured based on the user’s track through the use of sources, to say that source description is enriched with tags from the tagging history.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
