ABSTRACT
The widescale use of machine learning algorithms to drive decision-making has highlighted the critical importance of ensuring the interpretability of such models in order to engender trust in their output. The state-of-the-art recommendation systems use black-box latent factor models that provide no explanation of why a recommendation has been made, as they abstract their decision processes to a high-dimensional latent space which is beyond the direct comprehension of humans. We propose a novel approach for extracting explanations from latent factor recommendation systems by training association rules on the output of a matrix factorisation black-box model. By taking advantage of the interpretable structure of association rules, we demonstrate that predictive accuracy of the recommendation model can be maintained whilst yielding explanations with high fidelity to the black-box model on a unique industry dataset. Our approach mitigates the accuracy-interpretability trade-off whilst avoiding the need to sacrifice flexibility or use external data sources. We also contribute to the ill-defined problem of evaluating interpretability.
- Behnoush Abdollahi and Olfa Nasraoui. 2016. Explainable Matrix Factorization for Collaborative Filtering. WWW (Companion Volume) (2016), 5--6. Google ScholarDigital Library
- Behnoush Abdollahi and Olfa Nasraoui. 2016. Explainable Restricted Boltzmann Machines for Collaborative Filtering. 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016) Whi (2016). arXiv:1606.07129Google Scholar
- Behnoush Abdollahi and Olfa Nasraoui. 2017. Using Explainability for Constrained Matrix Factorization. Proceedings of the Eleventh ACM Conference on Recommender Systems - RecSys '17 (2017), 79--83. Google ScholarDigital Library
- Rakesh Agrawal, Tomasz Imielinski, and Arun Swami. 1993. Mining Association in Large Databases. Proceedings of the 1993 ACM SIGMOD international conference on Management of data - SIGMOD '93 (1993), 207--216. Google ScholarDigital Library
- Rakesh Agrawal and Ramakrishnan Srikant. 1994. Fast Algorithms for Mining Association Rules in Large Databases. Proceedings of the 20th International Conference on Very Large Data Bases (1994), 487--499. Google ScholarDigital Library
- Marharyta Aleksandrova, Armelle Brun, Anne Boyer, and Oleg Chertov. 2014. What about interpreting features in matrix factorization-based recommender systems as users? CEUR Workshop Proceedings 1210, 1 (2014).Google Scholar
- David Baehrens and T Schroeter. 2010. How to explain individual classification decisions. The Journal of Machine Learning Research 11 (2010), 1803--1831. arXiv:0912.1128 http://dl.acm.org/citation.cfm?id=1859912 Google ScholarDigital Library
- Bart Baesens, Rudy Setiono, Christophe Mues, and Jan Vanthienen. 2003. Using neural network rule extraction and decision tables for credit-risk evaluation. Management science 49, 3 (2003), 312--329. Google ScholarDigital Library
- Yang Bao, Fang Hui, and Jie Zhang. 2014. TopicMF: Simultaneously Exploiting Ratings and Reviews for Recommendation. Aaai (2014), 2--8. http://www.ntu. edu.sg/home/zhangj/paper/aaai14-bao.pdf Google ScholarDigital Library
- Nahla Barakat and Andrew P. Bradley. 2010. Rule extraction from support vector machines: A review. Neurocomputing 74, 1--3 (2010), 178--190. Google ScholarDigital Library
- R Caruana, H Kangarloo, J D Dionisio, U Sinha, and D Johnson. 1999. Case-based explanation of non-case-based learning methods. Proceedings. AMIA Symposium (1999), 212--5. http://www.ncbi.nlm.nih.gov/pubmed/10566351{%}0Ahttp://www. pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC2232607Google Scholar
- Finale Doshi-Velez and Been Kim. 2017. Towards A Rigorous Science of Interpretable Machine Learning. arXiv Ml (2017), 1--13. arXiv:arXiv:1702.08608v2 https://arxiv.org/pdf/1702.08608.pdfGoogle Scholar
- Jerome H. Friedman and Nicholas I. Fisher. 1999. Bump Hunting in High- Dimensional Data. Statics and Computing 9 (1999), 123--143. Google ScholarDigital Library
- Bryce Goodman and Seth Flaxman. 2016. EU regulations on algorithmic decisionmaking and a "right to explanation". 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016) Whi (2016), 26--30. arXiv:1606.08813 http://arxiv.org/abs/1606.08813Google Scholar
- Prem K. Gopalan, Laurent Charlin, and David Blei. 2014. Content-based recommendations with Poisson factorization. NIPS Advances in Neural Information Processing Systems (2014), 3176--3184. http://papers.nips.cc/paper/ 5360-content-based-recommendations-with-poisson-factorization Google ScholarDigital Library
- F Maxwell Harper and Joseph A Konstan. 2015. The MovieLens Datasets: History and Context. ACM Trans. Interact. Intell. Syst. 5, 4 (2015), 19:1--19:19. Google ScholarDigital Library
- John R Hauser, Olivier Toubia, Theodoros Evgeniou, Rene Befurt, and Daria Dzyabura. 2010. Disjunctions of Conjunctions, Cognitive Simplicity, and Consideration Sets. Journal of Marketing Research 47, 3 (2010), 485--496.Google ScholarCross Ref
- Reinhard Heckel, Michail Vlachos, Thomas Parnell, and Celestine Duenner. 2017. Scalable and interpretable product recommendations via overlapping co-clustering. Proceedings - International Conference on Data Engineering (2017), 1033--1044. arXiv:1604.02071Google ScholarCross Ref
- Jonathan L Herlocker, Joseph a Konstan, and John Riedl. 2000. Explaining collaborative filtering recommendations. Proceedings of the 2000 ACM conference on Computer supported cooperative work (2000), 241--250. arXiv:48 Google ScholarDigital Library
- Thomas Hofmann. 2004. Latent semantic models for collaborative filtering. ACM Transactions on Information Systems 22, 1 (2004), 89--115. Google ScholarDigital Library
- Yifan Hu, Chris Volinsky, and Yehuda Koren. 2008. Collaborative filtering for implicit feedback datasets. Proceedings - IEEE International Conference on Data Mining, ICDM (2008), 263--272. Google ScholarDigital Library
- Pang Wei Koh and Percy Liang. 2017. Understanding Black-box Predictions via Influence Functions. arXiv preprint arXiv:1703.04730 (2017). arXiv:1703.04730 http://arxiv.org/abs/1703.04730Google ScholarDigital Library
- Yehuda Koren. 2009. Collaborative filtering with temporal dynamics. Proc. of KDD '09 (2009), 447--456. Google ScholarDigital Library
- Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix factorization techniques for recommender systems. Computer 42, 8 (2009), 30--37. arXiv:ISSN 0018--9162 Google ScholarDigital Library
- Jiwei Li, Will Monroe, and Dan Jurafsky. 2016. Understanding Neural Networks through Representation Erasure. Arxiv (2016). arXiv:1612.08220 http://arxiv.org/ abs/1612.08220Google Scholar
- George A. Miller. 1956. The magical number seven, plus or minus two: some limits on our capacity for processing information. Psychological Review 63, 2 (1956), 81--97. arXiv:arXiv:1011.1669v3Google ScholarCross Ref
- Yu Mochizuki. 2016--. apyori: PyPI package for Python. (2016--). https://pypi. org/project/apyori/ {Online; accessed 20/05/2018}.Google Scholar
- Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. Model-Agnostic Interpretability of Machine Learning. ICML Workshop on Human Interpretability in Machine Learning Whi (2016). arXiv:1606.05386Google Scholar
- Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. "Why Should I Trust You?". Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '16 (2016), 1135--1144. arXiv:1602.04938Google ScholarDigital Library
- Sebastian Riedel, Limin Yao, Andrew McCallum, and Benjamin M. Marlin. 2013. Relation Extraction with Matrix Factorization and Universal Schemas. Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies June (2013), 74--84. http://www.aclweb.org/anthology/N13--1008Google Scholar
- Marco Rossetti, Fabio Stella, and Markus Zanker. 2013. Towards explaining latent factors with topic models in collaborative recommender systems. Proceedings - International Workshop on Database and Expert Systems Applications, DEXA (2013), 162--167. Google ScholarDigital Library
- Ruslan Salakhutdinov, Andriy Mnih, and Geoffrey Hinton. 2007. Restricted Boltzmann machines for collaborative filtering. Proceedings of the 24th international conference on Machine learning - ICML '07 (2007), 791--798. Google ScholarDigital Library
- Ivan Sanchez, Tim Rocktaschel, Sebastian Riedel, and Sameer Singh. 2015. Towards Extracting Faithful and Descriptive Representations of Latent Variable Models. AAAI Spring Symposium on Knowledge Representation and Reasoning 3 (2015), 35--38.Google Scholar
- Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl. 2000. Analysis of recommendation algorithms for e-commerce. Proceedings of the 2nd ACM conference on Electronic commerce - EC '00 (2000), 158--167. arXiv:117 Google ScholarDigital Library
- Tong Wang, Cynthia Rudin, Finale Doshi-Velez, Yimin Liu, Erica Klampfl, and Perry Macneille. 2015. Bayesian Or's of And's for Interpretable Classification with Application to Context Aware Recommender Systems. (2015), 1--40. arXiv:1504.07614Google Scholar
- Yongfeng Zhang, Guokun Lai, Min Zhang, Yi Zhang, Yiqun Liu, and Shaoping Ma. 2014. Explicit factor models for explainable recommendation based on phrase-level sentiment analysis. Proceedings of the 37th international ACM SIGIR conference on Research &development in information retrieval - SIGIR '14 (2014), 83--92. Google ScholarDigital Library
Index Terms
Explanation Mining: Post Hoc Interpretability of Latent Factor Models for Recommendation Systems
Recommendations
Using Explainability for Constrained Matrix Factorization
RecSys '17: Proceedings of the Eleventh ACM Conference on Recommender SystemsAccurate model-based Collaborative Filtering (CF) approaches, such as Matrix Factorization (MF), tend to be black-box machine learning models that lack interpretability and do not provide a straightforward explanation for their outputs. Yet explanations ...
Mining changes in association rules: a fuzzy approach
Association rule mining is concerned with the discovery of interesting association relationships hidden in databases. Existing algorithms typically assume that data characteristics are stable over time. Their main focus is therefore to mine association ...
A new approach based on association rules to add explainability to time series forecasting models
AbstractMachine learning and deep learning have become the most useful and powerful tools in the last years to mine information from large datasets. Despite the successful application to many research fields, it is widely known that some of these ...
Highlights- Novel methodology focused on visual explainability for time series forecasting.
- Use of association rules as an agnostic approach for adding explainability.
- Visual representation of the rules for a better interpretation of the ...
Comments