Dempster-Shafer Theory: Development of a New Software Implementation and Creation of a Benchmark

Authors

DOI:

https://doi.org/10.32603/2071-2340-2026-2-57-73

Keywords:

Dempster-Shafer theory, benchmarking, performance profiling, Python, open-source libraries, reproducibility of experiments

Abstract

The paper presents an open platform for reproducible testing and comparison of implementations of the core Dempster-Shafer algorithms. The platform includes a test data generator, a unified adapter interface for libraries, and a multi-level profiling system (CPU, memory, line-by-line source code analysis). An experimental comparison of four Python libraries was conducted: the authors' own implementation DSha, as well as the open-source dstpy, dstz, and pyds. The results demonstrate that DSha outperforms its counterparts across a range of metrics: it provides a speedup of 2 to 8.7 times over the full 4-step process while maintaining the most balanced memory consumption. It was found that pyds does not support the critical steps of discounting and Yager's rule, and dstz shows a performance drop (up to 8.7 times slower) at the combination steps due to inefficient set operations. Line-by-line source code analysis identified operations on set/frozenset as the main bottleneck across all implementations, indicating directions for further optimization. All experimental artifacts and source code are available in an open repository.

Author Biographies

  • Vladimir Parkhomenko, Peter the Great St. Petersburg Polytechnic University

    Senior Lecturer, Institute of Computer Science and Cybersecurity, SPbPU, vladimir.parkhomenko@spbstu.ru

  • Fedor Novikov, Peter the Great St. Petersburg Polytechnic University

    Dr. Sci. (Tech.), Professor, Institute of Physics and Mechanics, SPbPU, novikov_fa@spbstu.ru

  • Egor Plehanov, Peter the Great St. Petersburg Polytechnic University

    Master's Degree student, Institute of Computer Science and Cybersecurity, SPbPU, plehanov.es@edu.spbstu.ru

References

A. Dempster, "Upper and lower probabilities induced by a multivalued mapping," Annals of Mathematical Statistics, vol. 38, no. 2, pp. 325-339, 1967, doi: 10.1214/aoms/1177698950.

G. Shafer, A Mathematical Theory of Evidence. Princeton, NJ, USA: Princeton University Press, 1976.

reineking, "Библиотека pyds," GitHub. [Online]. Available: https://github.com/reineking/pyds.git

zttxtech, "dstz library", GitHub. [Online]. Available: https://github.com/ztxtech/dstz.git

geekabel, "dst-py library", GitHub. [Online]. Available: https://github.com/geekabel/dst-py.git

B. Cheung, "Dempster-shafer theory for classification using python," Blog post, Aug. 2020. [Online]. Available: https://bennycheung.github.io/dempster-shafer-theory-for-classification

E. S. Plehanov and V. A. Parkhomenko. DSha library source code. [Online]. Available: https://github.com/DM-groups/DSHa

L. V. Utkin, Risk Analysis and Decision Making under Incomplete Information. St. Petersburg, Russia: Nauka, 2007 (in Russian).

R. Yager, "On the dempster-shafer framework and new combination rules," Information Sciences, vol. 41, pp. 93-137, 1987, doi: 10.1016/0020-0255(87)90007-7.

V. Ivanov, N. Vinogradova, B. Palyukh, and A. Sotnikov, "Current trends and applications of dempster-shafer theory (review)," 2021, arXiv:2103.15592.

K. Zhao, L. Li, Z. Chen, R. Sun, G. Yuan, and J. Li, "A survey: Optimization and applications of evidence fusion algorithm based on dempster-shafer theory," Applied Soft Computing, vol. 126, p. 109075, 2022, doi: 10.1016/j.asoc.2022.109075.

K. Belmessous, F. Sebbak, M. Mataoui, M. Senouci, and W. Cherifi, "Dempster-shafer theory in recommender systems: A survey," International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, vol. 32, no. 05, 2024, doi: 10.1142/S0218488524500181.

A. Kaltsounidis and I. Karali, "Dempster-shafer theory: How constraint programming can help," in Proc. of the 13th International Conference on Scalable Uncertainty Management, 2020, pp. 373-386, doi: 10.1007/978-3-030-50143-3_27.

A. Paksoy and M. Gokturk, "Information fusion with dempster-shafer evidence theory for software defect prediction," Procedia Computer Science, vol. 3, pp. 600-605, 2011, doi: 10.1016/j.procs.2010.12.100.

E. S. Plehanov and V. A. Parkhomenko, DS-Benchmark project repository. [Online]. Available: https://github.com/DM-groups/DS-Benchmark

V. V. Podinovski, Ideas and Methods of Criteria Importance Theory in Multi-Criteria Decision Making Problems, Moscow: Nauka, 2019 (in Russian).

Published

2026-06-30

Issue

No. 2 (2026)

Section

Artificial intelligence and machine learning

License

Copyright (c) 2026 Владимир Пархоменко

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

[1]
V. Parkhomenko, F. Novikov, and E. Plehanov, “Dempster-Shafer Theory: Development of a New Software Implementation and Creation of a Benchmark”, Computer Tools in Education, no. 2, pp. 57–73, Jun. 2026, doi: 10.32603/2071-2340-2026-2-57-73.

Similar Articles

1-10 of 124

You may also start an advanced similarity search for this article.

Most read articles by the same author(s)