Learning strategies for an adaptive information retrieval system using neural networks

The results of an experimental investigation about the use of neural networks in associative adaptive information retrieval are presented. The learning and generalization capabilities of the backpropagation learning procedure are used to build up and employ application domain knowledge in the form of a sub-symbolic knowledge representation. The knowledge is acquired from examples of queries and relevant documents of the collection. In the tests reported, three different learning strategies are introduced and analyzed. Their results in terms of learning and generalization of the application domain knowledge are studied from an information retrieval point of view. The retrieval performance is studied and compared with that obtained by a traditional retrieval approach