Patent documents are an ample source of innovative, technical and commercial knowledge. Thus patent analysis has been considered a useful vehicle for R&D management and knowledge sharing. However, the number of patents in recent years has expanded noticeably. It is a challenge to interpret the knowledge contents and their relationship between patents, because many different types of expression may be presented in patents. Therefore, various clustering methods, e.g., partitioning clustering (K-means, K-medoids) or competitive clustering (Self-Organizing Maps, Adaptive Resonance Theory), have been developed to elucidate the characteristic of expression pattern. Nevertheless, patent documents can be partitioned into a number of clusters, where each cluster represents the documents in a certain area. The traditional clustering algorithms do not allow an object to belong to multiple clusters. In this research, we develop a novel non-exhaustive overlapping partitioning clustering (OPC) algorithm, a type of fuzzy partitioning approach, to the patent documents to overcome the exclusive clustering methods. The proposed algorithm considers the dissimilarity among cluster centers and an object can simultaneously belong to multiple clusters if the distances from this object to the cluster centers are no more than the given threshold value. Finally, the experiment uses radio frequency identification (RFID) technology related patent documents to evaluate the performance of the proposed clustering method, which combine non-exhaustive OPC algorithm with fuzzy adaptive resonance theory (ART) for effictive patent clustering.