Market-Basket Analysis Using Agglomerative Hierarchical Approach for Clustering a Retail Items

With the advent of data mining technology, cluster analysis of items is frequently done in supermarkets and in other large-scale retail sectors. Clustering of items has been a popular tool for identification of different groups of items where appropriate programs and techniques in data mining like Market-Basket analysis have been defined for each group separately with maximum effectiveness and return. For example, items frequently purchased together are placed in one place in the shelf of a retail store. There are various algorithms used for clustering. Hierarchical algorithms find successive clusters using previously established clusters. These algorithms usually are either agglomerative (bottom-up) or divisive (top-down). The paper presents the Market-Basket Analysis using Agglomerative (Bottom-up) hierarchical approach for clustering a retail items. Agglomerative hierarchical clustering creates a hierarchy of clusters which are represented in a tree structure called a Dendorogram. In agglomerative hierarchical clustering, dendrograms are developed based on the concept of distance between the entities or, groups of entities. . The clustering will done in such a way that the Purpose of Market-Basket Analysis will achieve.