Do News Consumers Want Explanations for Personalized News Rankings

@inproceedings{Hoeve2017DoNC,
  title={Do News Consumers Want Explanations for Personalized News Rankings},
  author={Maartje ter Hoeve and Mathieu Heruer and Daan Odijk and Anne Schuth and Martijn Spitters and M. de Rijke},
  year={2017}
}
This paper states the case for the principle of minimal necessary data: If two recommender algorithms achieve the same effectiveness, the better algorithm is the one that requires less user data. Applying this principle involves carrying out training data requirements analysis, which we argue should be adopted as best practice for the development and evaluation of recommender algorithms. We take the position that responsible recommendation is recommendation that serves the people whose data it… 

Figures and Tables from this paper

Action-Triggering Recommenders: Uplift Optimization and Persuasive Explanation
TLDR
This work applies a causal inference framework to estimate the average uplift for the offline evaluation of recommenders and proposes a new explanation style using context, which makes users imagine situations of items' usage and motivate them to purchase the items.
Democratizing algorithmic news recommenders: how to materialize voice in a technologically saturated media ecosystem
TLDR
This paper presents and argues for algorithmic recommender personae as a way for people to take more control over the algorithms that curate people's news provision by thinking about and enabling possibilities to express voice.
Do Users Appreciate Explanations of Recommendations? An Analysis in the Movie Domain
TLDR
In this paper, insights into users’ needs regarding the inclusion of explanations in a movie recommender system are provided and experimental results show that users do not always appreciate explanations.
Why does my model fail?: contrastive local explanations for retail forecasting
TLDR
It is shown that users are able to answer objective questions about the model's predictions with overall 81.1% accuracy when provided with these contrastive explanations, and that users who saw MC-BRP explanations understand why the model makes large errors in predictions significantly more than users in the control group.
The Need for Identifying Ways to Monetize Personalization and Recommendation
TLDR
The paper concludes with a discussion on the need for the research community to take multi-stakeholder interests into account in the design and evaluation of adaptive systems to foresee unwanted effects, such as online filter bubbles, and to pro-actively find strategies to prevent them.
Mining information interaction behavior : Academic papers and enterprise emails
TLDR
This thesis uncovers how academic searchers interact with information objects, and focuses on how users read their enterprise emails, and characterize user reading time, which improves understanding of user behavior on email platforms.
Machine Learning Explanations to Prevent Overtrust in Fake News Detection
TLDR
The research investigates the effects of an Explainable AI assistant embedded in news review platforms for combating the propagation of fake news and designs a news reviewing and sharing interface, creates a dataset of news stories, and trains four interpretable fake news detection algorithms.
Contrastive Explanations for Large Errors in Retail Forecasting Predictions through Monte Carlo Simulations
TLDR
The study shows that users are able to answer objective questions about the model’s predictions with overall 81.7% accuracy when provided with contrastive explanations, and that users who saw MC-BRP explanations understand why the model makes large errors in predictions significantly more than users in the control group.
Contrastive Explanations for Large Errors in Retail Forecasting Predictions through Monte Carlo Simulations
TLDR
It is shown that users are able to answer objective questions about the model's predictions with overall 81.7% accuracy when provided with these contrastive explanations, and that users who saw MC-BRP explanations understand why the model makes large errors in predictions significantly more than users in the control group.
Explainable Recommendations in Intelligent Systems: Delivery Methods, Modalities and Risks
TLDR
It is revealed that explanations delivery to end-users is mostly designed to be along with the recommendation in a push and pull styles while archiving explanations for later accountability and traceability is still limited.
...
...

References

SHOWING 1-10 OF 50 REFERENCES
Explaining collaborative filtering recommendations
TLDR
This paper presents experimental evidence that shows that providing explanations can improve the acceptance of ACF systems, and presents a model for explanations based on the user's conceptual model of the recommendation process.
Personalized news recommendation based on click behavior
TLDR
This research presents the content-based recommendation mechanism which uses learned user profiles with an existing collaborative filtering mechanism to generate personalized news recommendations in Google News and demonstrates that the hybrid method improves the quality of news recommendation and increases traffic to the site.
Explaining Recommendations: Satisfaction vs. Promotion
TLDR
This work presents two new methods for explaining recommendations of contentbased and/or collaborative systems and experimentally shows that they actually improve user’s estimation of item quality.
Designing and Evaluating Explanations for Recommender Systems
TLDR
This chapter gives an overview of the area of explanations in recommender systems, and approaches the literature from the angle of evaluation: that is, what makes an explanation “good”, and suggest guidelines as how to best evaluate this.
Personalized news recommendation based on implicit feedback
This paper presents a personalized news recommendation system that combines effective ways of understanding new articles with novel ways of modelling evolving user interest profiles to deliver
Tagsplanations: explaining recommendations using tags
TLDR
This paper develops novel algorithms for estimating tag relevance and tag preference, and conducts a user study exploring the roles of tag relevanceand tag preference in promoting effective tagsplanations.
Reviews, Reputation, and Revenue: The Case of Yelp.Com
Do online consumer reviews affect restaurant demand? I investigate this question using a novel dataset combining reviews from the website Yelp.com and restaurant data from the Washington State
Content-boosted collaborative filtering for improved recommendations
TLDR
This paper presents an elegant and effective framework for combining content and collaboration, which uses a content-based predictor to enhance existing user data, and then provides personalized suggestions through collaborative filtering.
A Survey of Explanations in Recommender Systems
  • N. TintarevJ. Masthoff
  • Computer Science
    2007 IEEE 23rd International Conference on Data Engineering Workshop
  • 2007
This paper provides a comprehensive review of explanations in recommender systems. We highlight seven possible advantages of an explanation facility, and describe how existing measures can be used to
...
...