A Bibliographic View on Constrained Clustering
This paper provides a bibliographic overview for researchers in constrained clustering, highlighting gaps like the lack of large comparison experiments, but it is incremental as it synthesizes existing literature without proposing new methods.
The authors analyzed about 3,000 documents on constrained clustering, compiling a detailed bibliography of 183 papers to identify general trends and sub-topics using Pareto analysis based on citation count and publication year, finding a notable lack of large comparison experiments and noting that applications studies, deep learning, active learning, and ensemble learning were most abundant recently.
A keyword search on constrained clustering on Web-of-Science returned just under 3,000 documents. We ran automatic analyses of those, and compiled our own bibliography of 183 papers which we analysed in more detail based on their topic and experimental study, if any. This paper presents general trends of the area and its sub-topics by Pareto analysis, using citation count and year of publication. We list available software and analyse the experimental sections of our reference collection. We found a notable lack of large comparison experiments. Among the topics we reviewed, applications studies were most abundant recently, alongside deep learning, active learning and ensemble learning.