Explainability Methods for Identifying Root-Cause of SLA Violation Prediction in 5G Network
The DOI matches its Crossref record; 23 of the 24 references it cites were found in a registry or an index.
Abstract
Artificial Intelligence (AI) is implemented in various applications of telecommunication domain, ranging from managing the network, controlling a specific hardware function, preventing a failure, or troubleshooting a problem till automating the network slice management in 5G. The greater levels of autonomy increase the need for explainability of the decisions made by AI so that humans can understand them (e.g. the underlying data evidence and causal reasoning) consequently enabling trust. This paper presents first, the application of multiple global and local explainability methods with the main purpose to analyze the root-cause of Service Level Agreement violation prediction in a 5G network slicing setup by identifying important features contributing to the decision. Second, it performs a comparative analysis of the applied methods to analyze explainability of the predicted violation. Further, the global explainability results are validated using statistical Causal Dataframe method in order to improve the identified cause of the problem and thus validating the explainability.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · cited 2×
“…paper: We present experimental setup for the 5G network slice, mainly focusing on transport domain, and apply multiple local (LIME [10]; Eli5 [11]; SHAP [12]) and global (SHAP [12]; PI [13]; XGBoost [14]) explainability methods to identify and explain the cause(s) of the SLA violation predictions.”
MethodGreedy function approximation: A gradient boosting machine
Friedman, J. H. (2001) · The Annals of Statistics · doi:10.1214/aos/1013203451 · Cited by 30,731
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707 · cited 3×
“Main contributions of the paper: We present experimental setup for the 5G network slice, mainly focusing on transport domain, and apply multiple local (LIME [10]; Eli5 [11]; SHAP [12]) and global (SHAP [12]; PI [13]; XGBoost [14]) explainability methods to identify and explain the cause(s) of the SLA violation predict…”
BackgroundMethodGauriat, C.-M., Pencolé, Y., Ribot, P., & Brouillet, G. (2024) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.4230/oasics.dx.2024.27 · Cited by 12,586
“Why Should I Trust You?”: Explaining the Predictions of Any Classifier
Ribeiro, M., Singh, S., & Guestrin, C. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations · doi:10.18653/v1/n16-3020 · Cited by 6,682
Permutation importance: a corrected feature importance measure
Altmann, A., Toloşi, L., Sander, O., & Lengauer, T. (2010) · Bioinformatics · doi:10.1093/bioinformatics/btq134 · Cited by 2,843 · cited 3×
“Another model agnostic explainer method is Permutation Importance (PI) [13] where it only generates global explanation of the system.”
MethodRobins, J. (1986) · Mathematical Modelling · doi:10.1016/0270-0255(86)90088-6 · Cited by 2,451 · cited 1×
“It implements Robin G’s formula [26] to calculate the causality effect of a feature by controlling the confounding factors.”
MethodAnchors: High-Precision Model-Agnostic Explanations
Ribeiro, M. T., Singh, S., & Guestrin, C. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11491 · Cited by 2,162
Goldstein, A., Kapelner, A., Bleich, J., & Pitkin, E. (2015) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2014.907095 · Cited by 1,861 · cited 1×
“A local explanation shows the factors that affect a particular decision [8]–[12] while global explanation shows the general characteristic of the predictor [12]–[17].”
BackgroundVisualizing the effects of predictor variables in black box supervised learning models
Apley, D. W., & Zhu, J. (2020) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/rssb.12377 · Cited by 1,535 · cited 1×
“A local explanation shows the factors that affect a particular decision [8]–[12] while global explanation shows the general characteristic of the predictor [12]–[17].”
BackgroundExplainable artificial intelligence: A survey
Dosilovic, F. K., Brcic, M., & Hlupic, N. (2018) · 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO) · doi:10.23919/mipro.2018.8400040 · Cited by 1,147 · cited 2×
“Several eXplainable AI (XAI) methods are available to analyze the decision made by the black box models [5].”
MethodInterpretable Counterfactual Explanations Guided by Prototypes
Van Looveren, A., & Klaise, J. (2021) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-86520-7_40 · Cited by 346
Explainable Artificial Intelligence for 6G: Improving Trust between Human and Machine
Guo, W. (2020) · IEEE Communications Magazine · doi:10.1109/mcom.001.2000050 · Cited by 253 · cited 1×
“Guo [24] focuses on trust and presents explainability for only physical and MAC layers in 6G.”
BackgroundCellular Traffic Prediction and Classification: A Comparative Evaluation of LSTM and ARIMA
Azari, A., Papapetrou, P., Denic, S. Z., & Peters, G. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-33778-0_11 · Cited by 129
Automatic Root Cause Analysis for LTE Networks Based on Unsupervised Techniques
Gómez-Andrades, A., Muñoz, P., Serrano, I., & Barco, R. (2015) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2015.2431742 · Cited by 118 · cited 1×
“One way to analyze a root cause automatically is by implementing self-organizing map (SOM) and data clustering in 5G network as done in [19].”
MethodWhen Machine Learning Meets Wireless Cellular Networks: Deployment, Challenges, and Applications
Challita, U., Rydén, H., & Tullberg, H. (2020) · IEEE Communications Magazine · doi:10.1109/mcom.001.1900664 · Cited by 87 · cited 1×
“Guo [24] focuses on trust and presents explainability for only physical and MAC layers in 6G. [6] emphasizes on the need of explain-ability for 5G especially to deliver critical services.”
BackgroundTowards Ubiquitous AI in 6G with Federated LearningPreprint
Xiao, Y. L., Shi, G., & Krunz, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.13563 · Cited by 50 · cited 2×
“There is a need to understand the automated decisions made using AI models [5]–[7].”
MethodRoot Cause Analysis Based on Temporal Analysis of Metrics Toward Self-Organizing 5G Networks
Muñoz, P., de-la-Bandera, I., Khatib, E. J., Gómez-Andrades, A., Serrano, I., & Barco, R. (2016) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2016.2586143 · Cited by 29 · cited 1×
“Another work [21] diagnoses a root cause using temporal analysis and measures the similarity between the time evolution of the operating network metrics with the stored faults.”
BackgroundAI for SLA management in programmable networks
Bendriss, J., Yahia, I. G. B., Chemouil, P., & Zeghlache, D. (2017) · HAL · Cited by 29
Adaptive Root Cause Analysis for Self-Healing in 5G Networks
Mfula, H., & Nurminen, J. K. (2017) · International Conference on High Performance Computing & Simulation (HPCS) · doi:10.1109/hpcs.2017.31 · Cited by 28 · cited 1×
“The work in [20] uses Bayesian Network classifier to calculate the probability of a cause which is capable in analyzing incomplete data or complex combinations of large dataset in self-healing 5G networks.”
MethodTowards automated service-oriented lifecycle management for 5G networks
Inam, R., Karapantelakis, A., Vandikas, K., Mokrushin, L., Feljan, A. V., & Fersman, E. (2015) · IEEE International Conference on Emerging Technologies and Factory Automation (ETFA) · doi:10.1109/etfa.2015.7301660 · Cited by 17 · cited 1×
“The concept of network slicing comes at a price of high complexity that stretches beyond human ability and needs automation [2].”
BackgroundSelf-citationSecured AI guided Architecture for D2D Systems of Massive MIMO deployed in 5G Networks
Vijay, A., & Umadevi, K. (2019) · International Conference on Trends in Electronics and Informatics (ICOEI) · doi:10.1109/icoei.2019.8862712 · Cited by 13 · cited 1×
“Another work in [25] presents only on security aspects of device-to-device architecture in 5G network.”
BackgroundA Machine Learning Based Management System for Network Services
Türk, Y., Zeydan, E., & Bilgin, Z. (2019) · IEEE International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob) · doi:10.1109/wimob.2019.8923572 · Cited by 4 · cited 1×
“Another management system applies ML methods to classify the network services state along with the root cause [23].”
Method
Research works printed but not linked: 3
- kelleher, 2018, Causality
- inam, 2017, 5G networks programmability for mission-critical applications, Ericsson Technology Review, 2
- 2015, 5G systems: Enabling industry and society transformations, Ericsson white paper, 3
Cited by
Explainable AI for 6G Use Cases: Technical Aspects and Research Challenges
Wang, S., Qureshi, M. A., Miralles-Pechuán, L., Huynh-The, T., Gadekallu, T. R., & Liyanage, M. (2024) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2024.3386872 · Cited by 170
Advancing 6G: Survey for Explainable AI on Communications and Network Slicing
Sun, H., Liu, Y., Al-Tahmeesschi, A., Nag, A., Soleimanpour-Moghadam, M., Canberk, B., Arslan, H., & Ahmadi, H. (2025) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2025.3534626 · Cited by 151
“According to [173], multiple methods, including LIME, SHAP, and XGBoost, are utilized to pinpoint the significant features contributing to SLA violation predictions.”
InfluentialLee, J., Solat, F., Kim, T. Y., & Poor, H. V. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3352910 · Cited by 151
“The study of [206] provides a comparative analysis of various XAI methods, such as XGBoost and SHapley Additive eXplanation (SHAP), to identify the root cause of SLA violations.”
BackgroundMethodBrik, B., Chergui, H., Zanzi, L., Devoti, F., Ksentini, A., Siddiqui, M. S., Costa‐Pérez, X., & Verikoukis, C. V. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3510543 · Cited by 142
“XAI would also assess the trustworthiness of AI functions via various metrics (see II-B), which is part of their SLA [37], [187].”
Nascita, A., Aceto, G., Ciuonzo, D., Montieri, A., Persico, V., & Pescapè, A. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3504955 · Cited by 98
“Morichetta et al. [165] and Dethise et al. [166] focus on predicting video quality, while Terra et al. [167] address the prediction of SLA violations.”
Influential
Metadata
- Authors
- Terra, A., Inam, R., Baskaran, S., Batista, P., Burdick, I., & Fersman, E.
- Affiliations
- Ericsson (Sweden) (SE)
- Ericsson (United States) (US)
- Venue
- GLOBECOM 2020 - 2020 IEEE Global Communications Conference
- Publisher
- IEEE
- Published
- 2020-12
- Pages
- 1-7
- ISSN
- 1930-529X (ISSN-L)
- OpenAlex
- W3123482879
- Cited by
- 50
- References
- 24
- Citations by year
- 2026: 2 · 2025: 13 · 2024: 14 · 2023: 9 · 2022: 10 · 2021: 2
- FWCI
- 2.01
- Type
- conference paper
- Language
- en
- Open access
- closed
- Licence
- Locations
- Globecom (publishedVersion)
- Funders
- Knut and Alice Wallenberg Foundation
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Computer Networks and Communications › Software-Defined Networks and 5G
- Physical Sciences › Computer Science › Computer Networks and Communications › Software System Performance and Reliability
- Keywords
- explainable AI, root cause analysis, 5G network slicing, feature importance, telecommunications network management
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.