Enhancing semi-supervised clustering

Abstract

Semi-supervised clustering employs limited supervision in the form of labeled instances or pairwise instance constraints to aid unsupervised clustering and often significantly improves the clustering performance. Despite the vast amount of expert knowledge spent on this problem, most existing work is not designed for handling high-dimensional sparse data. This paper thus fills this crucial void by developing a Semi-supervised Clustering method based on spheRical K-mEans via fEature projectioN (SCREEN). Specifically, we formulate the problem of constraint-guided feature projection, which can be nicely integrated with semi-supervised clustering algorithms and has the ability to effectively reduce data dimension. Indeed, our experimental results on several real-world data sets show that the SCREEN method can effectively deal with high-dimensional data and provides an appealing clustering performance.

A Survey On: Content Based Image Retrieval Systems Using Clustering Techniques For Large Data sets

“…Ref. [51] presented a semi-supervised clustering method base on spherical K-Means via feature projection which is tailored for handling sparse high dimensional data. They first formulated constraint-guided feature projection then applied the constraint spherical K-Means algorithm to cluster data with reduced dimension.…”

Network‐based semisupervised clustering

“…Moreover, Bair 8 evidences that traditional supervised classification methods are not useful when it is necessary to analyze a subset of labeled data. To solve the above mentioned problems of unsupervised clustering methods, several algorithms have been proposed to enhance clustering quality by employing supervision approaches 13,14 and, more precisely, by identifying the new class of semisupervised methods, that has recently become an important research topic 10,15‐18 . Semisupervised clustering is a variant of the traditional clustering paradigms.…”

“…In semisupervised clustering two different kinds of prior knowledge are considered: label information and pairwise constraints (or instance‐level constraints ) 18,23‐25 . Sometimes it is possible to distinguish between partial labels (a small subset of labeled instances) and instance‐level constraints (constraints characterizing pairs of data points) 25 .…”

Semi-Supervised Clustering Based on Exemplars Constraints

“…Moreover, she proposed a typical constrained kmeans algorithm based on these constraints. These advances led to many semi-supervised clustering methods based on pairwise constraints, such as constrained complete-link [4], constrained expectation-maximization (EM) [5], HMRFKmeans [6], MPCKmeans [7], kernel methods [8]- [13], matrix factorization-based methods [14], and constraint projection [15]- [18]. Kulis et al [8] reported a kernel algorithm to minimize a semi-supervised clustering objective function.…”