Informatics and Applications
2026, Volume 20, Issue 3, pp 87-95
METHODS FOR GENERATING METRICS ON OBJECT SETS IN THE CONTEXT OF TOPOLOGICAL DATA ANALYSIS THEORY. PART 2. EXPERIMENTAL GENERATION OF METRICS ON OBJECTS IN THE CONTEXT OF METRIC ALGORITHMS FOR NUMERICAL FORECASTING
Abstract
In the first part of this work, the main theoretical directions for generating problem-oriented metrics on sets of objects (pQ-metrics) using metrics on features (pL-metrics) were systematized. In this paper, criteria for tuning PQ-metrics are obtained using concepts of compactness. Computational experiments were conducted on 1000 independent data samples with 5000 pQ-metrics synthesized according to 20 proposed approaches for numerical forecasting by fc-nearest neighbor algorithms. Computational experiments demonstrated that in 95% cases, pQ-metrics of three types were effective: (i) those based on synthetic numerical features (61%); (ii) vector- based with pairwise matching (22%); and (iii) noncommutative metrics based on pL-distance arrays (12%). Importantly, pQ-metrics based on synthetic numerical features provided the best result for 83% of the datasets, which makes this approach the most promising.
[+] References (20)
- Zhuravlev, Yu. I. 1977. Correct algebras over sets of incorrect (heuristic) algorithms. I. Cybernetics 13(4):483-494.
- Zhuravlev, Yu. I. 1978. Ob algebraicheskom podkhode k resheniyu zadach raspoznavaniya i klassifikatsii [On the algebraic approach to solving recognition and classification tasks]. Problemy kibernetiki [Problems of Cybernetics] 33:5-68.
- Rudakov, K. V. 1988. Symmetric and functional constraints for classification algorithms. Soviet Mathematics Doklady 36(3):428-431.
- Torshin, I. Yu. 2026. Metody porozhdeniya metrik na mnozhestvakh ob"ektov v kontekste topologicheskoy teorii analiza dannykh. Chast' 1. Metriki na osnove rasstoyaniy mezhdu znacheniyami priznakov [Methods for generating metrics on object sets in the context of theory of topological data analysis. Part 1. Metrics based on distances between feature values]. Informatika i ee primeneniya - Inform. Appl. 20(1):39-44. doi: 10.14357/ 19922264260105. EDN: ZTDPJO.
- Kolmogorov, A. N., and S. V. Fomin. 1999. Elements of the theory of functions and functional analysis. New York, NY: Dover Publications. 448 p.
- Torshin, I. Yu. 2025. Metrizatsiya diskretnykh topologicheskikh prostranstv v kontekste teorii reshetok. Chast' 1. O normal'nosti prostranstv [Metrization of discrete topological spaces in the context of lattice theory. Part 1. On the normality of spaces]. Informatika i ee primeneniya - Inform. Appl. 19(1):82-88. doi: 10.14357/19922264250111. EDN: CAWKMO.
- Torshin, I. Yu., and K. V Rudakov. 2016. On metric spaces arising during formalization of recognition and classification problems. Part 1: Properties of compactness. Pattern Recognition Image Analysis 26(2):274-284. doi: 10.1134/S1054661816020255. EDN: WAMFRD.
- Deza, E. I., and M. M. Deza. 2006. Dictionary of distances. Amsterdam, Netherlands: North-Holland, Elsevier. 412 p. doi: 10.1016/B978-0-44452087-6.X5000-8.
- Hastie, T, R. Tibshirani, and J. H. Friedman. 2009. The elements of statistical learning: Data mining, inference, and prediction. New York, NY: Springer. 745 p. doi: 10.1007/978-0-387-84858-7.
- Hu, L., M. Huang, and S. Ke. 2016. The distance function effect on fc-nearest neighbor classification for medical datasets. SpringerPlus 5(1): 1304. 9p. doi: 10.1186/s40064- 016-2941-7.
- Adeniyi, D., Z. Wei, and Y. Yongquan. 2015. Automated web usage data mining and recommendation system using K-Nearest Neighbor (kNN) classification method. Applied Computing Informatics 12(1):90-108. doi: 10.1016/j.aci.2014.10.001.
- Cover, T., and P. Hart. 1967. Nearest neighbor pattern classification. IEEE T. Inform. Theory 13(1):21-27. doi: 10.1109/TIT.1967.1053964.
- Lopes, N., and B. Ribeiro. 2015. On the impact of distance metrics in instance-based learning algorithms. Pattern recognition and image analysis. Eds. R. Paredes, J. Cardoso, and X. Pardo. Lecture notes in computer science ser. Cham: Springer. 9117:48-56. doi: 10.1007/978-3- 319-19390-8_6.
- Mulak, P, and N. Talhar. 2015. Analysis of distance measures using k-nearest neighbor algorithm on KDD dataset. Int. J. Science Research 4(7):2101-2104.
- Edelsbrunner, H., and G. Osang. 2021. The multi-cover persistence of Euclidean balls. Discrete Comput. Geom. 65(4):1296-1313. doi: 10.1007/s00454-021-00281-9.
- Reani, Y., and O. Bobrowski. 2026. Morse theory for the kNN distance function. Discrete Comput. Geom. 75(2):491-510. doi: 10.1007/s00454-025-00795-6.
- Zou, J., and H. Li. 2022. Precise marketing of e-commerce products based on KNN algorithm. Comput. Intel. Neu- rosc. 2022:4966439. 12 p. doi: 10.1155/2022/4966439.
- Halder, R. K., M. N. Uddin, and M. A. Uddin. 2024. Enhancing K-nearest neighbor algorithm: A comprehensive
review and performance analysis of modifications. Big Data 11:113. 55 p. doi: 10.1186/s40537-024-00973-y.
- Abu Alfeilat, H.A., A.B.A. Hassanat, O. Lasassmeh, A. Tarawneh, M. Alhasanat, H. Eyal Salman, and V. Prasath. 2019. Effects of distance measure choice on K-nearest neighbor classifier performance: A review. Big Data 7(4):221-248. doi: 10.1089/big.2018.0175.
- Torshin, I. Yu. 2024. O porozhdenii sinteticheskikh priznakov na osnove opornykh tsepey i proizvol'nykh metrik v ramkakh topologicheskogo podkhoda k analizu dannykh. Chast' 1. Vklyuchenie v formalizm empiricheskikh funktsiy rasstoyaniya [On the generation of synthetic features based on support chains and arbitrary metrics within a topological approach to data analysis. Part 1. Inclusion of empirical distance functions into the formalism]. Informatika i ee Primeneniya - Inform. Appl. 18(1):71-77. doi: 10.14357/19922264240110. EDN: RIVOXR.
[+] About this article
Title
METHODS FOR GENERATING METRICS ON OBJECT SETS IN THE CONTEXT OF TOPOLOGICAL DATA ANALYSIS THEORY. PART 2. EXPERIMENTAL GENERATION OF METRICS ON OBJECTS IN THE CONTEXT OF METRIC ALGORITHMS FOR NUMERICAL FORECASTING
Journal
Informatics and Applications
2026, Volume 20, Issue 3, pp 87-95
Cover Date
2026-30-09
DOI
10.14357/19922264260308
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
topological data analysis; distance functions on objects; algebraic approach; feature value analysis theory
Authors
I. Yu. Torshin
Author Affiliations
 Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|