News Recommenders: Real-Time, Real-Life Experiences

Doychev, Doychin; Rafter, Rachael; Lawlor, Aonghus; Smyth, Barry

doi:10.1007/978-3-319-20267-9_28

Cited by 5 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consumption Behavior News items are often consumed anonymously and mostly without explicit user profiles (Doychev et al 2015 ; Sottocornola et al 2018 ). Though this problem can be mitigated by considering implicit signals like click patterns, reading time spent on an item, browsing and navigational patterns (Ilievski and Roy 2013 ; Trevisiol et al 2014 ), these implicit signals may sometimes be wrongly interpreted as an indicator of user’s appreciation or interests.…”

Section: Characteristics Of News Domainmentioning

confidence: 99%

“…They are based on the popularity of news items in terms of clickthrough rate, or social ties on social network sites. A traditional method of including popularity in an NRS is to simply count the total number of visits on news articles (Doychev et al 2015 ). However, calculating popularity based on top-N articles is prone to amplification (popularity bias or temporal bias), which is caused by exclusively selecting top-N articles while overlooking the good (N + 1)th candidate article (s).…”

Section: Major Challenges In News Recommender Systems and Conventional Solutionsmentioning

confidence: 99%

“…Duplication Similar content appears at multiple locations (URLs) from different news sources (Doychev et al 2015 ; Okura et al 2017 ; Robindro et al 2017 ). This can affect the ranking of news articles and is likely to bore the readers with repeated recommendations.…”

Section: Major Challenges In News Recommender Systems and Conventional Solutionsmentioning

confidence: 99%

“…The traditional statistical methods such as TF-IDF or BoW techniques based on content features are used to recommend similar news articles to the target user (Doychev et al 2015 ). But similar news articles are often repeated in the sense that they refer to the same news story presented in different ways from different publishers.…”

Section: Major Challenges In News Recommender Systems and Conventional Solutionsmentioning

confidence: 99%

See 3 more Smart Citations

News recommender system: a review of recent progress, challenges, and opportunities

2021

View full text Add to dashboard Cite

Nowadays, more and more news readers read news online where they have access to millions of news articles from multiple sources. In order to help users find the right and relevant content, news recommender systems (NRS) are developed to relieve the information overload problem and suggest news items that might be of interest for the news readers. In this paper, we highlight the major challenges faced by the NRS and identify the possible solutions from the state-of-the-art. Our discussion is divided into two parts. In the first part, we present an overview of the recommendation solutions, datasets, evaluation criteria beyond accuracy and recommendation platforms being used in the NRS. We also talk about two popular classes of models that have been successfully used in recent years. In the second part, we focus on the deep neural networks as solutions to build the NRS. Different from previous surveys, we study the effects of news recommendations on user behaviors and try to suggest possible remedies to mitigate those effects. By providing the state-ofthe-art knowledge, this survey can help researchers and professional practitioners have a better understanding of the recent developments in news recommendation algorithms. In addition, this survey sheds light on the potential new directions.

show abstract

Section: Characteristics Of News Domainmentioning

confidence: 99%

Section: Major Challenges In News Recommender Systems and Conventional Solutionsmentioning

confidence: 99%

Section: Major Challenges In News Recommender Systems and Conventional Solutionsmentioning

confidence: 99%

Section: Major Challenges In News Recommender Systems and Conventional Solutionsmentioning

confidence: 99%

See 2 more Smart Citations

News recommender system: a review of recent progress, challenges, and opportunities

2021

View full text Add to dashboard Cite

show abstract

“…While this might be suitable in some domains where the content does not change too often (e.g., in movie databases), it fails in more dynamic domains where items continuously emerge and extend collections, and where existing items become less and less relevant. Examples include news, microblog, or advertisement recommendations [3,4,5,7,6,10] where content comes in the form of a constant stream of data.…”

Section: Recommendations Of Streamed Datamentioning

confidence: 99%