Tagsplanations: explaining recommendations using tags

By Jesse Vig, Shilad Sen, John Riedl. 2009

366 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Safety, Robustness & Interpretability, Recommender 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

In Information Systems

In Safety, Robustness & Interpretability

In Recommender Systems

Citations in the corpus

RippleNet: Propagating User Preferences on the Knowledge Graph for Recommender Systems (2018, background): “ reveal why a user might like a particular item, which helps improve their acceptance or satisfaction of recommendations and increase trust in RS. The explanations are usually based on community tags [29], social networks [23], aspect [2], and phrase sentiment [44] Since RippleNet explores users’ in”
Knowledge-aware Autoencoders for Explainable Recommender Systems (2018, background): “Since the explanation may be decoupled from the recommendation process, a distinction between transparency and justification has to be made [20].”
Explainable Recommendation: A Survey and New Perspectives (2018, methodology): “Wang et al. [2018b] further generalized previous MF-based explainable recommendation models by multi-task learning with tensor factorization.”
A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines (2018, background): “Taggingdigital resources, such as movies and products [148] or even questions in forums [149], helps the users in ﬁnding the information they are interested in, hence the importance in designing tag recommendation systems.”
Explainable Outfit Recommendation with Joint Outfit Matching and Comment Generation (2018, methodology): “[15] propose an explainable recommendation method that uses community tags to generate explanations.”
Using Explainability for Constrained Matrix Factorization (2017, methodology): “In all styles, the data sources employed in the recommendation task, may be di erent from the data sources used in generating the explanation [4, 5, 9, 18], leading to explanation generation modules that are separate from the recommender system.”
Explainable Restricted Boltzmann Machines for Collaborative Filtering (2016, methodology): “Also, data sources employed in the recommendation task, may be diﬀerent from the data sources used in generating the explanation (Vig et al., 2009; Herlocker et al., 2000; Symeonidis et al., 2008; Bilgic & Mooney, 2005; Billsus & Pazzani, 1999).”
Explanation-based Ranking in Opinionated Recommender Systems (2018, methodology): “Keyword approaches were further developed by [13, 11] to justify recommendations as: “Item A is suggested because it contains feature X and Y that are also included in items B, C, and D, which you have also liked.”
Tagommenders: connecting users to items through tags (2009, neutral): “[32].”
User Preferences for Hybrid Explanations (2017, background): “To better understand user preferences, research has compared explanation styles [2], combined di erent explanation styles [24], and considered dimensions such as personalization [25, 26], tags [30], ranking [17], and natural language presentations [5].”
A Hybrid Explanations Framework for Collaborative Filtering Recommender Systems (2014, methodology): “In this work we present a “black box” explanations framework based on tags (e.g., as in [7]) in order to compute personalized explanations in the form of “We recommend Shrek based on your history of watching humorous computer animated films featuring imaginary storyline fitting a family outing”.”
Explanatory Debugging: Supporting End-User Debugging of Machine-Learned Programs (2010, background): “Explanations of learned programs’ logic have taken a variety of forms, such as relating user actions and the resulting predictions [2], detailing why a program made a particular prediction [14], or explaining how an outcome resulted from user actions [8, 23, 24].”
A systematic review and taxonomy of explanations in decision support and recommender systems (2017, background): “However, this deserves further investigation, because extracting explanations from today’s decision inference methods is becoming increasingly complex due to the increasing complexity of widely adopted ML algorithms, which may be even confidential, as argued by both Zanker andNinaus (2010) andVig et”
ReEL: Review Aware Explanation of Location Recommendation (2018, result): “Contrary to that, some of the studies [17, 18, 35, 36, 40, 41, 43, 57] have already claimed the user persuasiveness due to explainability in real-world systems.”
Do social explanations work?: studying and modeling the effects of social explanations in recommender systems (2013, background): “Using user-generated tags, based on their popularity and relevance, is another source of social information that has also been studied for explanation [27].”