Autors: Mecheva, T. A.
Title: Mecheva, T., 2026, Application of the K-Means Algorithm for Clustering and Analysis of Spatio-Temporal Transport Data, Proc. Int. Sci. Conf. Mechatronic, Eco- and Energy Saving Systems and Technologies (SMEEST), Gabrovo: University Publishing House "V. Aprilov", pp. 39–44.
Keywords: K-Means; clustering; traffic patterns; intelligent transport

Abstract: Clustering temporal traffic data can improve road safety and efficiency in Intelligent Transportation Systems. The aim of the study is to analyse spatio-temporal transport data in order to extract traffic characteristics on the section of St. Petersburg Boulevard in Plovdiv. The present study is part of a larger group of studies that apply conventional methods for clustering data. The data are initially grouped into 24-hour time series for each direction, missing values are filled, and then K-means clustering of speed and number of vehicles passed is applied. The number of clusters is determined by the Silhouette score, and the results are visualized via MDS transformation. The results show that traffic flow in the direction of the Ring Road exhibits greater variability - the traffic volume fluctuates within a wider range throughout the day compared to the traffic heading toward the city Center.

References

  1. Mecheva, T. (2026) „Application of the K-Means Algorithm for Clustering and Analysis of Spatio-Temporal Transport Data“, International Scientific Conference "Mechatronic, Eco- and Energy Saving Systems and Technologies" (SMEEST), Plovdiv, Bulgaria, 19 June 2026. Gabrovo: University Publishing House "V. Aprilov", pp. 39–44.

Issue

International Scientific Conference "Mechatronic, Eco- and Energy Saving Systems and Technologies" (SMEEST), pp. 39–44, 2026, Bulgaria, University Publishing House "V. Aprilov" - Gabrovo, ISBN ISSN: 2815-4924

Full text of the publication

Вид: публикация в национален форум с межд. уч., публикация в реферирано издание