Linked open data-based explanations for transparent recommender systems

Musto, Cataldo; Narducci, Fedelucio; Lops, Pasquale; Gemmis, Marco de; Semeraro, Giovanni

doi:10.1016/j.ijhcs.2018.03.003

Cited by 72 publications

(49 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One advantage of the explicit semantics of Linked Data is that they can also be used to optimize specific measures like diversity (Musto et al 2017a). Furthermore, Linked Data can serve as a basis to generate explanations and to thereby increase recommendation transparency (Musto et al 2019). Overall, data from the LOD cloud were successfully applied for recommendations in several domains, and various opportunities exist to connect LOD information with other types of side information, as shown in Oramas et al (2017) for music recommendation.…”

Section: Data-related Trendsmentioning

confidence: 99%

Trends in content-based recommendation

Lops

Jannach

Musto

et al. 2019

User Model User-Adap Inter

Self Cite

156

View full text Add to dashboard Cite

show abstract

Section: Data-related Trendsmentioning

confidence: 99%

Trends in content-based recommendation

Lops

Jannach

Musto

et al. 2019

User Model User-Adap Inter

Self Cite

156

View full text Add to dashboard Cite

show abstract

“…The cold-start problem is resolved in this work, and they stated that their model outperformed other baseline methods in terms of producing a lower error rate. A study conducted by Musto et al [26] shed light on the significance of natural language explanations in recommender systems and how linked open data can empower them by linking the user's previously preferred items and items' attributes to the new recommendations. The explanation mechanism is based on the notion that descriptive properties that describe the items that the user liked in the past can serve as explanations for the outputs of the recommender system.…”

Section: B Explanations In Black Box Recommender Systemsmentioning

confidence: 99%

“…No participant responded with the strongly unsatisfied answer option. Figures 24,25,26,27,28,29,and 30, show the responses of all participants to the demographic questions in Table 12. The answers for these questions were optional.…”

Section: Figure 21mentioning

confidence: 99%

Mining Semantic Knowledge Graphs to Add Explainability to Black Box Recommender Systems

2019

View full text Add to dashboard Cite

Recommender systems are being increasingly used to predict the preferences of users on online platforms and recommend relevant options that help them cope with information overload. In particular, modern model-based collaborative filtering algorithms, such as latent factor models, are considered stateof-the-art in recommendation systems. Unfortunately, these black box systems lack transparency, as they provide little information about the reasoning behind their predictions. White box systems, in contrast, can, by nature, easily generate explanations. However, their predictions are less accurate than sophisticated black box models. Recent research has demonstrated that explanations are an essential component in bringing the powerful predictions of big data and machine learning methods to a mass audience without compromising trust. Explanations can take a variety of formats, depending on the recommendation domain and the machine learning model used to make predictions. The objective of this work is to build a recommender system that can generate both accurate predictions and semantically rich explanations that justify the predictions. We propose a novel approach to build an explanation generation mechanism into a latent factor-based black box recommendation model. The designed model is trained to learn to make predictions that are accompanied by explanations that are automatically mined from the semantic web. Our evaluation experiments, which carefully study the trade-offs between the quality of predictions and explanations, show that our proposed approach succeeds in producing explainable predictions without a significant sacrifice in prediction accuracy.

show abstract

“…The cold-start problem is resolved in this work, and they stated that their model outperformed other baseline methods in terms of producing a lower error rate. A study conducted by [7] shed light on the significance of natural language explanations in recommender systems and how linked open data can empower them by linking the user's previously preferred items and items' attributes to the new recommendations. The explanation mechanism is based on the notion that descriptive properties that describe the items that the user liked in the past can serve as explanations for the outputs of the recommender system.…”

Section: Explanation Styles and Related Approachesmentioning

confidence: 99%

“…Hybrid CF system using MF that is accurate and transparent enough for users to accept recommendations by generating explanations through semantic web technologies. TasteWeights[3] is an interactive hybrid recommender system designed for the music domain7 . Several sources of information, such as Twitter, Facebook, and Wikipedia, are utilized as a data source for the recommendation process.…”

mentioning

confidence: 99%

An explainable recommender system based on semantically-aware matrix factorization.

Alshammari¹

View full text Add to dashboard Cite

show abstract

Linked open data-based explanations for transparent recommender systems

Cited by 72 publications

References 40 publications

Trends in content-based recommendation

Trends in content-based recommendation

Mining Semantic Knowledge Graphs to Add Explainability to Black Box Recommender Systems

An explainable recommender system based on semantically-aware matrix factorization.

Contact Info

Product

Resources

About