Spectral Co-Clustering is an implementation of Co-Clustering that models the input data as a bipartite graph, where the observations and features form two sets of nodes that are connected by edges. Thus, it is an extension of the spectral clustering setup that uses a subset of both the observations and features in the clustering process. A possible use case for spectral Co-Clustering is text classification, where one set of nodes might consist of words found across a set of documents that are linked to the documents in the second set of nodes, where the edges indicate that word w appears within document d.
What is Spectral co-clustering?
Help us improve this post by suggesting in comments below:
– modifications to the text, and infographics
– video resources that offer clear explanations for this question
– code snippets and case studies relevant to this concept
– online blogs, and research publications that are a “must read” on this topic
Partner Ad
