2020/03/07 by Rosana Veroneze, Veroneze, Rosana, Fernando J. Von Zuben +1
Computer Science · #Data Management and Algorithms #Data Mining Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Data Classification #Rough Sets and Fuzzy Logic
paper · pdf · doi:10.48550/arxiv.2003.04726
openalex publication_date 2020/03/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
This paper further extends RIn-CloseCVC, a biclustering algorithm capable of\nperforming an efficient, complete, correct and non-redundant enumeration of\nmaximal biclusters with constant values on columns in numerical datasets. By\navoiding a priori partitioning and itemization of the dataset, RIn-CloseCVC\nimplements an online partitioning, which is demonstrated here to guide to more\ninformative biclustering results. The improved algorithm is called\nRIn-CloseCVC3, keeps those attractive properties of RIn-CloseCVC, as formally\nproved here, and is characterized by: a drastic reduction in memory usage; a\nconsistent gain in runtime; additional ability to handle datasets with missing\nvalues; and additional ability to operate with attributes characterized by\ndistinct distributions or even mixed data types. The experimental results\ninclude synthetic and real-world datasets used to perform scalability and\nsensitivity analyses. As a practical case study, a parsimonious set of relevant\nand interpretable mixed-attribute-type rules is obtained in the context of\nsupervised descriptive pattern mining.\n