Soft Clustering

• Published in: Introduction To Pattern Recognition And Machine Learning Vol. 5; pp. 1 - 2
• Main Authors: Murty M Narasimha, Devi V Susheela
• Format: Book Chapter
• Language: English
• Published: Singapore World Scientific 2015; World Scientific Publishing Company; Co-Published with Indian Institute of Science (IISc), Bangalore, India
• Series: IISc Lecture Notes Series
• Subjects: Computer Hardware Engineering; Embedded Systems & Programming; General References; Pattern recognition; Software Engineering
• ISBN: 9814335452; 9789814335454; 9814656275; 9789814335461; 9814335460; 9789814656276
• DOI: 10.1142/9789814335461_0008

Clustering has been one of the most popular tools in data mining. Even though hard clustering has been popularly studied and traditionally used, there are several important applications where soft partitioning is essential. Some of these applications include text mining and social networks. Soft partitioning is concerned with assigning a document in text mining or a person in a social network to more than one cluster. For example, the same document may belong to both sports and politics. In several cases softness has to be appropriately interpreted to make a hard decision; such a process permits us to delay the decision making. So, one of the major characteristics of softness is in delaying decision making as far as possible. Another notion is to acknowledge the assignment of a pattern to more than one category...