The objective is to add an efficient implementation of the following algorithm into the ELKI data mining framework (this requires good Java knowledge):

Ding, C., & He, X. (2004, March). K-nearest-neighbor consistency in data clustering: incorporating local information into global optimization. In Proceedings of the 2004 ACM symposium on Applied computing (pp. 584-589).

and study the result quality compared to existing algorithms in ELKI. It is required that the implementation follows the object-oriented design patterns of ELKI, i.e., reusability and modularity. This likely requires the method to be implemented in two parts: the knn-based evaluation measure as well as the k-means-based optimization algorithm. The algorithm must be tested with appropriate unit tests.

This is only suitable as a Bachelor thesis topic.

Good Java programming skills

Good understanding of data mining algorithms

Good statistical knowledge