Russian Federation
Russian Federation
Russian Federation
The article compares three clustering algorithms - K-Means, DBSCAN and hierarchical clustering - to solve the problem of segmentation of users of digital services. Based on experiments with synthetic and real data, their effectiveness, noise tolerance, speed of operation, and interpretability of the results are evaluated. It is shown that the choice of an algorithm depends on the data structure and business goals, and high-quality data preprocessing is a key success factor.
user segmentation, cluster analysis, K-Means, DBSCAN, hierarchical clustering, unsupervised machine learning, behavioral analysis, data analysis, comparative analysis of algorithms
1. Volkova, E. E. Osobennosti povedeniya rossiyskih potrebiteley v vozraste 18-23 let na rynke cifrovogo muzykal'nogo kontenta / E. E. Volkova // Hypothesis. – 2018. – № 2(3). – S. 33-40. – EDN LIWTSS.
2. Prokopenko, N. Yu. Primenenie Loginom dlya optimizacii processov upravleniya tovarnymi zapasami predpriyatiy malogo i srednego biznesa / N. Yu. Prokopenko, D. V. Trishin // Mezhvuzovskiy sbornik statey laureatov konkursov : Sbornik statey / Nizhegorodskiy gosudarstvennyy arhitekturno-stroitel'nyy universitet. Tom VYPUSK 21. – Nizhniy Novgorod : Nizhegorodskiy gosudarstvennyy arhitekturno-stroitel'nyy universitet, 2021. – S. 216-222. – EDN TLNXXH.
3. Vityaev, E. E. Semanticheskiy veroyatnostnyy vyvod predskazaniy / E. E. Vityaev // Izvestiya Irkutskogo gosudarstvennogo universiteta. Seriya: Matematika. – 2017. – T. 21. – S. 33-50. – DOIhttps://doi.org/10.26516/1997-7670.2017.21.33. – EDN ZFOTBF.
4. Sumin, V. I. Osobennosti vybora chlenov ekspertnoy gruppy dlya analiza funkcionirovaniya slozhnoy organizacionnoy sistemy silovyh struktur / V. I. Sumin, A. S. Dubrovin, I. S. Kuscheva // Modelirovanie sistem i processov. – 2024. – T. 17, № 4. – S. 77-83. – DOIhttps://doi.org/10.12737/2219-0767-2024-17-4-77-83. – EDN ISTAML.
