Benchmarks for Data Clustering Algorithms in Data Science Kingston University in United Kingdom

This PhD opportunity at Kingston University focuses on advancing the field of data clustering within Data Science, particularly through the development and benchmarking of clustering algorithms for Big Data. Data Science challenges often involve processing and analysing vast, complex datasets, such as sensor data, social media, biological, and chemical data. The sheer volume and diversity of Big Data make retrieval, processing, and analysis both challenging and time-consuming. Clustering is a key unsupervised learning technique that helps unravel these challenges by categorising unlabelled data into meaningful groups, enabling the discovery of patterns and insights.

The project will investigate major paradigms in data clustering, including density-based, partition-based, hierarchical, parallel, and distributed approaches. The student will create benchmark datasets to systematically compare clustering algorithms, addressing the current gap in standardised evaluation. The research will extend existing benchmarks from 2D to 3D, increase the complexity from two clusters to multiple clusters, and diversify cluster shapes beyond circular forms. This work builds on previous research by the supervisor and incorporates new developments in the field.

Students will have the opportunity to apply the best clustering techniques to a problem domain of their choice, contributing to both theoretical and practical advancements. The expected outcomes include high-quality research publications and the development of a strong track record in Data Science and Machine Learning. The project is supported by the Faculty of Engineering, Computing and the Environment, offering access to a vibrant research community and resources.

Funding is available through the Kingston University Graduate School studentships competition for October 2026 entry. Details regarding tuition coverage and stipends can be found on the university's PhD Studentships webpage. Applicants should have a strong background in Computer Science, Data Science, Mathematics, or Statistics, with experience in data analysis and machine learning being highly desirable. Proficiency in programming and familiarity with big data tools will be advantageous. English language requirements must be met as per university standards.

The application deadline is March 4, 2026. Prospective candidates should review the application instructions on the Kingston University PhD Studentships and Faculty research webpages, prepare their materials, and submit their applications accordingly. This is an excellent opportunity for motivated students to contribute to cutting-edge research in data clustering and benchmarking, with the potential for significant impact in academia and industry.