Transaction Data Attached To The Problem Of Clustering Research

Clustering is the process of grouping the data into classes or clusters so that objects within a cluster have high similarity in comparison to one another, but are very dissimilar to objects in other clusters. Clustering analysis has been studied extensively, and many methods were found to solve various kinds of problems. Clustering algorithms can be divided into hard or fuzzy. A hard clustering algorithm allocates each pattern to a single cluster during its operation and in its output. A fuzzy clustering method assigns degrees of membership in several clusters to each input patterns, the degrees of membership are between 0 and 1. A fuzzy clustering can be converted to a hard clustering by assigning each pattern to the cluster with the largest measure of membership.But in some cases, a pattern can be allocated to more than one cluster. In this thesis we call it multi-subjected clustering. For numerical data, fuzzy clustering algorithms can be used to solve this kind of problems, but new algorithms need to be developed to solve the problem of multi-subjected clustering on transaction data or categorical data.With focus on transaction data, three algorithms were developed to solve multi-subjected clustering problems. There are frequent-items based algorithm, SLR-based algorithm and link-based algorithm. These algorithms can also be used on categorical data if the data were preprocessed.