SOMSE: A Neural Network Based Approach to Web Search Optimization

Conventional Web search engines return long lists of ranked documents that users are forced to sift through to find relevant documents. The notoriously low precision of Web search engines coupled with the ranked list presentation make it hard for users to find the information they are looking for. One of the fundamental issues of information retrieval is searching for compromises between precision and recall. It is generally desirable to have high precision and high recall, although in reality, increasing precision often means decreasing recall and vice versa. Developing retrieval techniques that will yield high recall and high precision is desirable. Unfortunately, such techniques would impose additional resource demands on the search engines. Search engines are under severe resource constraints and dedicating enough CPU time to each query might not be feasible. A more productive approach, however, seems to enhance post-processing of the retrieved set, such as providing links and semantic maps to retrieved results of a query. If such value-adding processes allow the user to easily identify relevant documents from a large retrieved set, queries that produce low precision/high recall results will become more acceptable. We propose improving the quality of Web search by combining meta-search and self-organizing maps. This can help users both in locating interesting documents more easily and in getting an overview of the retrieved document set.