Book chapter
Multimodal Clustering of Boolean Tensors on MapReduce: Experiments Revisited
This paper presents further development of distributed multimodal clustering. We introduce a new version of multimodal clustering algorithm for distributed processing in Apache Hadoop on computer clusters. Its implementation allows a user to conduct clustering on data with modality greater than two. We provide time and space complexity of the algorithm and justify its relevance. The algorithm is adapted for MapReduce distributed processing model. The program implemented by means of Apache Hadoop framework is able to perform parallel computing on thousands of nodes.