ABSTRACT
This chapter explores the clustering techniques in bioinformatics. The clustering algorithms are aptly titled advanced clustering techniques because they are based on the clustering techniques, but with natural extensions. Graph-based clustering is used to group similar vertices into one cluster such that the maximum number of edges connect within the cluster, and the minimum number of edges connect between the clusters. Adjacency-based measures can be performed by measuring the similarity between vertices using an adjacency matrix, which determines whether two vertices are similar or not by analyzing the overlap of the neighbors. Connectivity measures can be used to find the similarity between the vertices whether the vertices are in the same cluster or in different clusters. Cluster fitness can be determines using two approaches: density measure and cut-based measure. Higher-order mining is a data mining form in which derived data, statistical information, or patterns are the input instead of raw data.
