On July 2023, HS4U published its first scientific article on Mathematics (MDPI) entitled “Reducing Uncertainty and Increasing Confidence Unsupervised Learning”. Mathematics is a peer- reviewed, open access journal which provides forum for studies related to mathematics and is published semimonthly online by MDPI.

HS4U’s article presents the development of a novel algorithm for an unsupervised learning called RUN-ICON (Reduce Uncertainty and Increase CONfidence). The primary objective of the algorithm is to enhance the reliability and confidence of unsupervised clustering. RUN-ICON leverages the K-means++ methodology, a well-known approach for clustering data points. However, it distinguishes itself from existing K-means variants by introducing innovative metrics. Rather than relying solely on the Sum of Squared Errors (SSE) as a measure of cluster quality, RUN-ICON incorporates novel metrics such as the Clustering Dominance Index and Uncertainty. Furthermore, the algorithm features characteristics such as robustness, high-quality clustering, automation, and flexibility. Extensive testing on diverse data sets with varying characteristics has proven its capability to determine the optimal number of clusters under different scenarios.

The algorithm will soon be deployed in real-world scenarios, where it will undergo rigorous testing against data sets based on measurements and simulations, further proving its effectiveness.

To read the full scientific article, click here.