Applying connectionist models to information retrieval
Sally Jo Cunningham, Geoffrey Holmes, Jamie Littin, Russell Beale, Ian H. Witten · 1998
: Adaptive information retrieval (IR) systems based on connectionist architectures have captured the attention of researchers over the past decade. This paper provides a review of connectionist IR research, including the major models for connectionist document and query representation, techniques to enhance query re-formulation, dynamic document routing (information filtering), and connectionist techniques for document clustering. 1. Introduction Information retrieval (IR) systems provide populations of users with access to a large collection of stored information. These systems are concerned with the structure, analysis, organization, storage, and searching of such information. Collections have typically comprised text documents but are increasingly involving new data types such as pictures, audio, video, and so on. The main goal of an IR system is to extract from the files of documents those items that most closely correspond to requests from the user population. These systems...