The availability of effective clusters is crucial to many document retrieval systems. MacLeod's algorithm incorporates ART1-like training with static similarity thresholds to form effective document clusters and is well suited to massively parallel implementations [1]. This paper reports on a study of the effectiveness of clusters formed by the algorithm when several variable threshold functions are. . .
MacLeod, K. (2013). Performance in ART1-Like Document Clustering with Variable Similarity Thresholds. Proceedings of the Annual Conference of CAIS Actes Du congrès Annuel De l’ACSI. https://doi.org/10.29173/cais150