Abstract
Explainable Artificial Intelligence (XAI) has received a lot of attention over the past decade, with the proposal of many methods explaining black box classifiers such as neural networks. Despite the ubiquity of recommender systems in the digital world, only few researchers have attempted to explain their functioning, whereas one major obstacle to their use is the problem of societal acceptability and trustworthiness. Indeed, recommender systems direct user choices to a large extent and their impact is important as they give access to only a small part of the range of items (e.g., products and/or services), as the submerged part of the iceberg. Consequently, they limit access to other resources. The potentially negative effects of these systems have been pointed out as phenomena like echo chambers and winner-take-all effects, because the internal logic of these systems is to likely enclose the consumer in a “déjà vu” loop. Therefore, it is crucial to provide explanations of such recommender systems and to identify the user data that led the respective system to make the individual recommendations. This then makes it possible to evaluate recommender systems not only regarding their effectiveness (i.e., their capability to recommend an item that was actually chosen by the user), but also with respect to the diversity, relevance and timeliness of the active data used for the recommendation. In this paper, we propose a deep analysis of two state-of-the-art models learnt on four datasets based on the identification of the items or the sequences of items actively used by the models. Our proposed methods are based on subgroup discovery with different pattern languages (i.e., itemsets and sequences). Specifically, we provide interpretable explanations of the recommendations of the Top-N items, which are useful to compare different models. Ultimately, these can then be used to present simple and understandable patterns to explain the reasons behind a generated recommendation to the user.
Similar content being viewed by others
Notes
n is randomly chosen in the interval [5, 25].
NB: we removed duplicated perturbed sequences in a post-processing step.
References
Atzmueller M (2015) Subgroup discovery. WIREs Data Min Knowl Discov 5(1):35–49
Atzmueller M, Lemmerich F (2009) Fast subgroup discovery for continuous target concepts. In: Proceedings international symposium on methodologies for intelligent systems, vol 5722, LNCS. Springer, Berlin, pp 1–15
Atzmueller M, Puppe F (2006) SD-Map: a fast algorithm for exhaustive subgroup discovery. In: Proceedings of PKDD. Springer, pp 6–17
Bloemheuvel S, Kloepper B, van den Hoogen J, Atzmueller M (2019) Enhancing sequential pattern mining explainability with markov chain probabilities. In: Proceedings of Dutch-Belgian database day, Jheronimus Academy of Data Science, Den Bosch, Netherlands
Duivesteijn W, Thaele J (2014) Understanding where your classifier does (not) work: the SCaPE model class for EMM. In: Proceedings of ICDM. IEEE, pp 809–814
Falher GL, Gionis A, Mathioudakis M (2015) Where is the Soho of Rome? measures and algorithms for finding similar neighborhoods in cities. In: Proceedings of the ninth international conference on web and social media, ICWSM 2015, University of Oxford, Oxford, UK, May 26–29, 2015, pp 228–237
Fournier-Viger P, Lin JC-W, Kiran RU, Koh YS, Thomas R (2017) A survey of sequential pattern mining. Data Sci Pattern Recognit 1(1):54–77
Fürnkranz J, Kliegr T, Paulheim H (2020) On cognitive preferences and the plausibility of rule-based models. Mach Learn 109(4):853–898
Gulla JA, Zhang L, Liu P, Özgöbek O, Su X (2017) The adressa dataset for news recommendation. In: Proceedings of the international conference on web intelligence, WI ’17, New York, NY, USA. Association for Computing Machinery, pp 1042–1048
Han J, Pei J, Yin Y (2000) Mining frequent patterns without candidate generation. In: Proceedings of ACM SIGMOD international conference on management of data. ACM Press, pp 1–12
Harper FM, Konstan JA (2015) The movielens datasets: history and context. ACM TiiS 5(4):19:1-19:19
Henelius A, Puolamäki K, Boström H, Asker L, Papapetrou P (2014) A peek into the black box: exploring classifiers by randomization. Data Min Knowl Discov 28(5–6):1503–1529
Kang W, McAuley JJ (2018) Self-attentive sequential recommendation. In: Proceedings of ICDM. IEEE, pp 197–206
Kendall MG (1938) A new measure of rank correlation. Biometrika 30(1/2):81–93
Klösgen W (1996) Explora: a multipattern and multistrategy discovery assistant. In: Advances in knowledge discovery and data mining. AAAI, pp 249–271
Lemmerich F, Becker M, Atzmueller M (2012) Generic pattern trees for exhaustive exceptional model mining. In: Proceedings of ECML/PKDD. Springer, Berlin/Heidelberg, Germany
Lemmerich F, Atzmueller M, Puppe F (2016) Fast exhaustive subgroup discovery with numerical target concepts. Data Min Knowl Discov 30(3):711–762
Lonjarret C, Robardet C, Plantevit M, Auburtin R, Atzmueller M (2020) Why should I trust this item? Explaining the recommendations of any model. In: Webb GI, Zhang Z, Tseng VS, Williams G, Vlachos M, Cao L (eds) 7th IEEE international conference on data science and advanced analytics, DSAA 2020, Sydney, Australia, Oct 6–9, 2020. IEEE, pp 526–535
Lonjarret C, Auburtin R, Robardet C, Plantevit M (2021) Sequential recommendation with metric models based on frequent sequences. Data Min Knowl Discov 35(3):1087–1133
Mabroukeh NR, Ezeife CI (2010) A taxonomy of sequential pattern mining algorithms. ACM Comput Surv 43(1):3:1-3:41
Mandel DR (2007) Counterfactual and causal explanation: from early theoretical views to new frontiers. In: The psychology of counterfactual thinking. Routledge, pp 23–39
Mathonat R, Nurbakova D, Boulicaut J, Kaytoue M (2019) Seqscout: using a bandit model to discover interesting subgroups in labeled sequences. In: Proceedings of DSAA, pp 81–90
McAuley J, Targett C, Shi Q, van den Hengel A (2015) Image-based recommendations on styles and substitutes. In: Proceedings of SIGIR. ACM, pp 43–52
McSherry D (2005) Explanation in recommender systems. Artif Intell Rev 24(2):179–197
Mollenhauer D, Atzmueller M (2020) Sequential exceptional pattern discovery using pattern-growth: an extensible framework for interpretable machine learning on sequential data. In: International workshop on explainable and interpretable machine learning. CEUR-WS.org, vol 2796
Pei J, Han J, Mortazavi-Asl B, Pinto H, Chen Q, Dayal U, Hsu M (2001) Prefixspan: mining sequential patterns by prefix-projected growth. In: Young DC (ed) Proceedings of international conference on data engineering, Los Alamitos. IEEE, pp 215–224
Pei J, Han J, Mortazavi-Asl B, Wang J, Pinto H, Chen Q, Dayal U, Hsu M (2004) Mining sequential patterns by pattern-growth: the prefixspan approach. IEEE Trans Knowl Data Eng 16(11):1424–1440
Pope PE, Kolouri S, Rostami M, Martin CE, Hoffmann H (2019) Explainability methods for graph convolutional neural networks. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 10772–10781
Pu P, Chen L (2006) Trust building with explanation interfaces. In: Proceedings of IUI, pp 93–100
Ribeiro MT, Singh S, Guestrin C (2016) “Why should I trust you?” Explaining the predictions of any classifier. In: Proceedings of KDD, pp 1135–1144
Ribeiro MT, Singh S, Guestrin C (2018) Anchors: high-precision model-agnostic explanations. In: Proceedings of AAAI
Roth-Berghofer TR, Cassens J (2005) Mapping goals and kinds of explanations to the knowledge containers of case-based reasoning systems. In: Proceedings of ICCBR, number 3620 in LNAI. Springer, Berlin, pp 451–464
Roth-Berghofer T, Schulz S, Leake D, Bahls D (2007) Explanation-aware computing. AI Mag 28(4):122
Schank RC (1986) Explanation patterns: understanding mechanically and creatively. Lawrence Erlbaum Associates, Hillsdale
Sørmo F, Cassens J, Aamodt A (2005) Explanation in case-based reasoning: perspectives and goals. Artif Intell Rev 24(2):109–143
Tang J, Wang K (2018) Personalized top-n sequential recommendation via convolutional sequence embedding. In: Proceedings of the eleventh ACM international conference on web search and data mining, WSDM ’18, New York, NY, USA. ACM, pp 565–573
Tintarev N, Masthoff J (2011) Designing and evaluating explanations for recommender systems. In: Recommender systems handbook. Springer, pp 479–510
Tolomei G, Silvestri F, Haines A, Lalmas M (2017) Interpretable predictions of tree-based ensembles via actionable feature tweaking. In: Proceedings of KDD. ACM, pp 465–474
Wick MR, Thompson WB (1992) Reconstructive expert system explanation. Artif Intell 54(1–2):33–70
Wrobel S (1997) An algorithm for multi-relational discovery of subgroups. In: Proceedings of PKDD, number 1263 in LNCS, Berlin/Heidelberg, Germany. Springer, pp 78–87
Zaki MJ (2001) Spade: an efficient algorithm for mining frequent sequences. Mach Learn 42(1–2):31–60
Acknowledgements
This work was supported by the ACADEMICS grant of the IDEXLYON project of the University of Lyon, PIA operated by ANR-16-IDEX-0005. It also benefited from the financial support CAF AMERICA OPE2020-0041.
Author information
Authors and Affiliations
Corresponding author
Additional information
Responsible editor: Alipio Jorge.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
A Appendix
A Appendix
Hit rate (Hit@N):
where the indicator function \(\mathbb {1}(b)\) returns 1 if its argument b is True, 0 otherwise. \(R_{i}\) is the ranking of the ground-truth item i. The HIT@N function returns the average number of times the ground-truth item is ranked in the Top-N items. We compute HIT@5, HIT@10, HIT@25.
Normalized Discounted Cumulative Gain at position N(nDCG@N):
The NDCG@k is a position-aware metric which assigns larger weights to higher positions. We compute NDCG@5, NDCG@10, NDCG@25 and NDCG@50.
Area Under Curve (AUC):
This measure calculates how high the ground-truth items of each user has been ranked in average.
Precision at N (Precision@N):
Recall at N (Recall@N):
Mean Average Precision (MAP):
where \(rel(k) = 1\) if the kth item in TN belongs to GT, the ground Truth items of the user.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Iferroudjene, M., Lonjarret, C., Robardet, C. et al. Methods for explaining Top-N recommendations through subgroup discovery. Data Min Knowl Disc 37, 833–872 (2023). https://doi.org/10.1007/s10618-022-00897-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10618-022-00897-2