Using Explainability for Constrained Matrix Factorization

By Behnoush Abdollahi, Olfa Nasraoui. 2017

124 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Yongfeng Zhang
3. Xu Chen
4. Xiang Wang
5. Tat-seng Chua
6. Liqiang Nie
7. Fuli Feng
8. Xia Hu

In Information Systems

1. Xiangnan He
2. Yongfeng Zhang
3. Xu Chen
4. Xiang Wang
5. Tat-seng Chua
6. Liqiang Nie
7. Fuli Feng
8. Xing Xie

Citations in the corpus

• Explainable Recommendation: A Survey and New Perspectives (2018, background): “3: A comparison between relevant-item explanation (left) and relevant-user explanation (right) (Abdollahi and Nasraoui, 2017).”
• TEM: Tree-enhanced Embedding Model for Explainable Recommendation (2018, methodology): “The similarity-based methods [1, 2] present explanations as a list of most similar users or items.”
• The FacT: Taming Latent Factor Models for Explainability with Factorization Trees (2019, methodology): “However, the fidelity of explanation and the quality of recommendation have long been considered as irreconcilable [1]: one has to trade recommendation quality for explanation.”
• In2Rec: Influence-based Interpretable Recommendation (2019, methodology): “Obviously, In2Rec has significant improvements compared with EMF, especially on items with low item rating frequency.”
• A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “The approach from Abdollahi and Nasraoui (2016) also uses neighborhood style explanations. An explainability score with a range of zero to one is used to determine whether the user rated a product in a similar way that his neighborhood users did. If the value is zero, the user is not contributing to”
• Evaluating Explanation Without Ground Truth in Interpretable Machine Learning (2019, methodology): “ explanations generated from search-based methodologies (65), instead of those gradient- (66) or perturbation-based (67) methodologies. Typical examples for capability evaluation can be found in work (29,68) with the application to recommender systems, where both work employed explainability precisi”
• Explainable Recommendation via Multi-Task Learning in Opinionated Text Data (2018, background): “algorithms has attracted increasing attention in the research community. Zhang et al. [6, 38] combined techniques for phrase-level sentiment analysis with matrix factorization. Abdollahi and Nasraoui [1] introduced explainability as a constraint in factorization: the learnt latent factors for a user”
• Explanation Mining: Post Hoc Interpretability of Latent Factor Models for Recommendation Systems (2018, background): “Abdollahi and Nasraoui [3] extends [1] using a probabilistic formulation of EMF.”