Processing of chemical and textual databases using the ICL DAP and transputer networks

Discusses studies of the use of DAP for text retrieval and cluster analysis. Text retrieval has involved a serial file organisation using a nearest neighbour search system. Cluster analysis (or automatic classification) allows the identification of groups of similar objects in multidimensional data sets. Initial studies used nonhierarchical clustering, later experiments used fixed format data records such as chemical structure records. Database searching on transputer networks has mainly involved database parallelism