INFORMATION TECHNOLOGY FOR CONSTRUCTING EXPLANATIONS USING TEMPORALLY ORDERED INPUT DATA OF AN INTELLIGENT SYSTEM
DOI:
https://doi.org/10.20998/2079-0023.2026.01.10Keywords:
explainable artificial intelligence, temporal delay, temporal dependencies, causal dependencies, temporal graph neural network, information technologyAbstract
The subject of the article is the process of generating explanations for the decisions of intelligent systems whose input data are temporally ordered sequences of events with time delays. The aim of the work is to develop an approach to constructing explanations for the decisions of intelligent systems with temporally ordered input data that takes into account time delays in the input event sequences. To achieve this aim, the following tasks are addressed: to develop a method for processing temporal delays that includes a delay estimation component based on the cross‑correlation function and a time‑encoding component; to develop an information technology for constructing explanations for the decisions of an intelligent system based on the temporal order of input data, which integrates the developed method into a single pipeline for explanation construction and verification; to carry out experimental evaluation of the method and the information technology. A method for processing temporal delays in causal dependencies between nodes of a dynamic graph in a temporal graph neural network is proposed, which differs from known approaches by combining components of correlation‑based estimation of optimal temporal shift, phase‑shifted time encoding, and adaptive fusion of the obtained representations, thereby enabling the incorporation of causal dependencies into explanations through the estimation of time delays. An information technology for constructing explanations for the decisions of an intelligent system based on the temporal order of input data is proposed, which includes the stages of adaptive construction of temporal event graphs, building a temporal graph neural network with temporal delay processing, generation and subsequent verification of explanations based on temporal algebra, thus providing the formation of explanations that take into account changes in the order of the intelligent system’s input events. The experimental evaluation has confirmed that the temporal delay processing method adapts to deterministic, stochastic, and cyclic delays.
References
Arrieta A. B., Díaz-Rodríguez N., Del Ser J., Bennetot A., Tabik S., Barbado A., García S., Gil-López S., Molina D., Benjamins R., Chatila R., Herrera F. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion. 2020, vol. 58, pp. 82–115. DOI: 10.1016/j.inffus.2019.12.012.
Guidotti R., Monreale A., Ruggieri S., Turini F., Giannotti F., Pedreschi D. A Survey of Methods for Explaining Black Box Models. ACM Computing Surveys. 2018, vol. 51, no. 5, article 93, pp. 1–42. DOI: 10.1145/3236009.
Ying R., Gu R., Yu K., Bourgeois D., You J., Zitnik M., Leskovec J. GNNExplainer: Generating Explanations for Graph Neural Networks. Advances in Neural Information Processing Systems (NeurIPS 2019). 2019, vol. 32, pp. 9244–9255.
Sundararajan M., Taly A., Yan Q. Axiomatic Attribution for Deep Networks. Proceedings of the 34th International Conference on Machine Learning (ICML 2017). PMLR 70. 2017, pp. 3319–3328.
Ribeiro M. T., Singh S., Guestrin C. “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2016). 2016, pp. 1135–1144. DOI: 10.1145/2939672.2939778.
Xu D., Ruan C., Korpeoglu E., Kumar S., Achan K. Inductive Representation Learning on Temporal Graphs. Proceedings of the 8th International Conference on Learning Representations (ICLR 2020). 2020.
Rossi E., Chamberlain B., Frasca F., Eynard D., Monti F., Bronstein M. Temporal Graph Networks for Deep Learning on Dynamic Graphs. ICML 2020 Workshop on Graph Representation Learning and Beyond (GRL+). 2020. arXiv:2006.10637.
Pareja A., Domeniconi G., Chen J., Ma T., Suzumura T., Kanezashi H., Kaler T., Schardl T., Leiserson C. EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs. Proceedings of the AAAI Conference on Artificial Intelligence. 2020, vol. 34, no. 04, pp. 5363–5370. DOI: 10.1609/aaai.v34i04.5984.
Chala O. V. Pobudova temporalnykh pravyl dlia predstavlennia znan v informatsiinykh systemakh upravlinnia. Suchasni informatsiini systemy. 2018, vol. 2, no. 3, pp. 54–58. DOI: 10.20998/2522- 9052.2018.3.09. (In Ukr.)
Chala O. V. Model uzahalnenoho predstavlennia temporalnykh znan v intelektualnykh informatsiinykh systemakh upravlinnia. Suchasni informatsiini systemy. 2020, vol. 4, no. 2, pp. 30–35. DOI: 10.20998/2522-9052.2020.2.05. (In Ukr.)
Ismail Fawaz H., Forestier G., Weber J., Idoumghar L., Muller P.-A. Deep Learning for Time Series Classification: a Review. Data Mining and Knowledge Discovery. 2019, vol. 33, no. 4, pp. 917–963. DOI: 10.1007/s10618-019-00619-1.
Ye B., Yang S., Hu B., Zhang Z., He Y., Huang K., Zhou J., Fang Y. Gaia: Graph Neural Network with Temporal Shift Aware Attention for Gross Merchandise Value Forecast in E-commerce. Proceedings of the 38th IEEE International Conference on Data Engineering (ICDE 2022). 2022, pp. 1–13. DOI: 10.1109/ICDE53745.2022.00245.
Chalyi S. F., Kravchenko R. V. Metod adaptyvnoho vyboru intervaliv chasu dlia pobudovy hrafiv temporalnykh hrafovykh neironnykh merezh. Visnyk NTU «KhPI». Seriia: Systemnyi analiz, upravlinnia ta informatsiini tekhnolohii. 2025, no. 2 (12), pp. 4–11. DOI: 10.20998/2079-0023.2025.02.01. (In Ukr.)
Chalyi S. F., Kravchenko R. V. Hrafova neironna merezha dlia temporalno vporiadkovanykh danykh u zadachi pobudovy poiasnen v intelektualnii systemi. Suchasni informatsiini systemy. 2025, vol. 9, no. 2. DOI: 10.20998/2522-9052.2025.2.XX. (In Ukr.)
Chalyi S. F., Kravchenko R. V. Metod pobudovy temporalno uzhodzhenykh poiasnen v intelektualnykh systemakh na osnovi temporalnykh hrafovykh neironnykh merezh. Suchasni informatsiini systemy. 2025, vol. 9, no. 3. DOI: 10.20998/2522-9052.2025.3.XX. (In Ukr.)
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
