A Personalized and Context-Aware News Offer for Mobile Devices

Pessemier, Toon De; Vanhecke, Kris; Martens, Luc

doi:10.1007/978-3-319-30996-5_8

Cited by 4 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…by linking users from the same locality and offering similar recommendations) (Thurman and Schifferes, 2012). Because news consumption always occurs in a certain context, these factors have significant impact on users' decisions and preferences (de Pessemier et al, 2016).…”

Section: Ethical Dilemmas Of Algorithmic News Personalizationmentioning

confidence: 99%

News personalization for peace: how algorithmic recommendations can impact conflict coverage

Bastian

Makhortykh

Dobber

2019

IJCMA

View full text Add to dashboard Cite

Purpose The purpose of this paper is to develop a conceptual framework for assessing what are the possibilities and pitfalls of using algorithmic systems of news personalization – i.e. the tailoring of individualized news feeds based on users’ information preferences – for constructive conflict coverage in the context of peace journalism, a journalistic paradigm calling for more diversified and creative war reporting. Design/methodology/approach The paper provides a critical review of existing research on peace journalism and algorithmic news personalization, and analyzes the intersections between the two concepts. Specifically, it identifies recurring pitfalls of peace journalism based on empirical research on constructive conflict coverage and then introduces a conceptual framework for analyzing to what degree these pitfalls can be mediated – or worsened – through algorithmic system design. Findings The findings suggest that AI-driven distribution technologies can facilitate constructive war reporting, in particular by countering the effects of journalists’ self-censorship and by diversifying conflict coverage. The implementation of these goals, however, depends on multiple system design solutions, thus resonating with current calls for more responsible and value-sensitive algorithmic design in the domain of news media. Additionally, our observations emphasize the importance of developing new algorithmic literacies among journalists both to realize the positive potential of AI for promoting peace and to increase the awareness of possible negative impacts of new systems of content distribution. Originality/value The article particle is the first to provide a comprehensive conceptualization of the impact of new content distribution techniques on constructive conflict coverage in the context of peace journalism. It also offers a novel conceptual framing for assessing the impact of algorithmic news personalization on reporting traumatic and polarizing events, such as wars and violence.

show abstract

Section: Ethical Dilemmas Of Algorithmic News Personalizationmentioning

confidence: 99%

News personalization for peace: how algorithmic recommendations can impact conflict coverage

Bastian

Makhortykh

Dobber

2019

IJCMA

View full text Add to dashboard Cite

show abstract

“…Tidlige systemer brukte artikkelklikk som indikasjon på interesse, men noen klikker ved uhell og andre forlater artikkelen når de ser at innholdet ikke var som forventet (Epure et al, 2017). Mange nyere anbefalingsløsninger bruker en variant av lesetid for å utlede leserens interesse for artikkelen (De Pessemier et al, 2015. Ideen er at en leser trolig har interesse for artikkeltemaet dersom vedkommende tok seg tid til å lese artikkelen grundig og til siste slutt.…”

Section: Implisitte Evalueringerunclassified

Anbefaling av nyhetsinnhold i praksis

Svendsen¹,

Gulla²,

Frøland³

2019

NMT

View full text Add to dashboard Cite

Automatiserte anbefalinger av nyhetsinnhold brukes i dag på nettsidene til mange medieselskaper med hensikt å presentere leserne mer relevante nyheter og tilby bedre brukeropplevelser. Slike anbefalingsløsninger drar nytte av nye teknikker fra maskinlaering og stordataarkitekturer fra informatikkfaget. I denne artikkelen diskuterer vi om personlige nyhetsanbefalinger kan øke trafikken til nettaviser i en tid hvor mediebransjen er under press, samt bidra til å gi medieaktørene større brukerforståelse. I denne artikkelen belyser vi noen av utfordringene med nyhetsanbefalinger og diskuterer spesielt teknikker som kollaborativ filtrering og innholdsbaserte anbefalinger. Vi kommer også inn på evalueringsmetrikker, fordi tradisjonelle mål på presisjon og recall ikke alltid samsvarer med mediehusenes behov for å benytte løsningene strategisk for å bedre lønnsomheten. Polaris Media har over flere år eksperimentert med anbefalingsteknologi på sine nettaviser, og vi presenterer i artikkelen funn og innsikter fra denne utprøvingen. Noen hovedfunn er at anbefalingsløsninger kan øke klikkraten og lesetiden betydelig, revitalisere eldre nyhetsstoff slik at innhold får lengre levetid samt bidra til å forstå brukersegmentene bedre. Artikkelen posisjonerer seg innenfor en design science research-tilnaerming, hvor målsetningen er å forstå industriens behov og utvikle og prøve ut artefakter som møter disse behovene.

show abstract

“…Driven by this success, more application domains have adopted recommender systems to reduce the information overload by generating personalized suggestions. Also for recruitment scenarios, in which applicants search for suitable job offers, recommender systems are a useful tool for job candidates, recruiters, as well as platforms that connect both [12]. The mainstream approaches to recommender systems are classified into four categories:…”

Section: Introductionmentioning

confidence: 99%

Employment Recommendation System: A Review

Belsare¹,

Deshmukh²

2018

IJSEA

View full text Add to dashboard Cite

Enormous amounts of jobs are posted on the job websites on daily basis and large numbers of new resumes are also added to job websites daily. In such scenario it"s a very tough job to suggest matching jobs to the job applicants. A recommendation system can solve this problem to the great extent. A recommendation system has already been proved to be very effective in the area of Online shopping websites and Movie recommendation. Given a user, the goal of an employment recommendation system is to predict those job positions that are likely to be relevant to the user. An Employment recommendation system would suggest matching jobs to the users using matching, collaborative filtering and content based recommendation based on ranking.

show abstract

A Personalized and Context-Aware News Offer for Mobile Devices

Cited by 4 publications

References 26 publications

News personalization for peace: how algorithmic recommendations can impact conflict coverage

News personalization for peace: how algorithmic recommendations can impact conflict coverage

Anbefaling av nyhetsinnhold i praksis

Employment Recommendation System: A Review

Contact Info

Product

Resources

About