Emotional news recommender system

Parizi, Ali Hakimi; Kazemifard, Mohammad

doi:10.1109/cogsci.2015.7426666

Cited by 11 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CF is a class of recommender systems based on user feedback in the form of liked, disliked, ratings about items, etc. CF approach predicts and recommends items or products to users (or subset of users) having similar tastes, using predictive models based on the past ratings or likeness behavior of all the users [6], [32], [33], [78], [84], [91]. CF is a useful approach as compare to content-based approach where content analysis is difficult, e.g.…”

Section: A Collaborative Filtering Approachmentioning

confidence: 99%

“…In content-based filtering approach, users or items are considered as atomic units and better-personalized recommendation is being made knowing more about user or item. For example, demographic information of a user or item features like stars, director or genre of a movie [6], [78], [83], [91]. The conventional TF-IDF vector-space model has commonly adopted technique for content-based filtering approach used to select a set of candidate items.…”

Section: B Content-based Approachmentioning

confidence: 99%

“…Semantically rich domain knowledge models incorporation might produce good results for news recommendation beyond the traditional recommendation techniques. Semantic-based systems like contextualization [9], [91], context-dependent relevance [10], domain ontology [41] and other related studies briefly discussed below;…”

Section: B Semantic-based News Recommendationmentioning

confidence: 99%

“…News recommendation model based on the emotion of news and how the user feels about the news article is introduced by [91]. The model is used to recommend news that has positive impact on user's mood, utilizing both emotion and user's preferences.…”

Section: B Semantic-based News Recommendationmentioning

confidence: 99%

See 3 more Smart Citations

News Recommendation Systems - Accomplishments, Challenges & Future Directions

et al. 2020

View full text Add to dashboard Cite

News publishers have decreased disseminating news through conventional newspapers and have migrated to the use of digital means like websites and purpose-built mobile applications. It is observed that news recommendation systems can automatically process lengthy articles and identify similar articles for readers considering predefined criteria. The objectives of the current work are to identify and classify the challenges in news recommendation domain, to identify state-of-the-art approaches and classify on the application domain, to identify datasets used for evaluation and their sources, the evaluation approaches used and to highlight the challenges explicitly addressed. The literature is thoroughly studied over the time span of 2001-2019 and shortlisted 81 related studies, broadly classified into six categories and discussed. The analysis showed that 60% of news recommendation system adopted a hybrid approach, 66% studies little talk about datasets, and addresses a few challenges from a long list of challenges in the news domain. This article is the first in the field to draw a comprehensive big picture of news recommendation and explore different dimensions covered in the studies. The last section presents the future research opportunities that lead to improving the recommendation of news articles in the news domain.

show abstract

Section: A Collaborative Filtering Approachmentioning

confidence: 99%

Section: B Content-based Approachmentioning

confidence: 99%

Section: B Semantic-based News Recommendationmentioning

confidence: 99%

Section: B Semantic-based News Recommendationmentioning

confidence: 99%

See 2 more Smart Citations

News Recommendation Systems - Accomplishments, Challenges & Future Directions

et al. 2020

View full text Add to dashboard Cite

show abstract

“…An emerging and ever expanding approach to such personalisation is emotion recognition in which the mood or affective state of the user is approximated and then used to modify or adapt the system functionality or appearance [1]. This is particularly ever more apparent in recommender systems, such as [28], [31], [33] few but to give examples.…”

Section: Introductionmentioning

confidence: 99%

Class discovery from semi-structured EEG data for affective computing and personalisation

Ayesh

Arevalillo‐Herráez

Arnau-González

2017

2017 IEEE 16th International Conference on Cognitive Informatics &Amp; Cognitive Computing (ICCI*CC)

View full text Add to dashboard Cite

Many approaches to recognising emotions from metrical data such as EEG signals rely on identifying a very small number of classes and to train a classifier. The interpretation of these classes varies from a single emotion such as stress [24] to features of emotional model such as valence-arousal [4]. There are two major issues here. First classification approach limits the analysis of the data within the selected classes and is also highly dependent on training data/cycles, all of which limits generalisation. Second issue is that it does not explore the interrelationships between the data collected missing out on any correlations that could tell us interesting facts beyond emotional recognition. This second issue would be of particular interest to psychologists and medical professions. In this paper, we investigate the use of Self-Organizing Maps (SOM) in identifying clusters from EEG signals that could then be translated into classes. We start by training varying sizes of SOM with the EEG data provided in a public dataset (DEAP). The produced graphs showing Neighbour Distance, Sample Hits, Weight Position are analysed holistically to identify patterns in the structure. Following that, we have considered the groundtruth label provided in DEAP, in order to identify correlations between the label and the clustering produced by the SOM. The results show the potential of SOM for class discovery in this particular context. We conclude with a discussion on the implications of this work and the difficulties in evaluating the outcome.

show abstract

The Shortcomings of the Diversity Diet: Public Service Media, Algorithms and the Multiple Dimensions of Diversity

Sørensen

2022

Palgrave Global Media Policy and Business

View full text Add to dashboard Cite

Emotional news recommender system

Cited by 11 publications

References 18 publications

News Recommendation Systems - Accomplishments, Challenges & Future Directions

News Recommendation Systems - Accomplishments, Challenges & Future Directions

Class discovery from semi-structured EEG data for affective computing and personalisation

The Shortcomings of the Diversity Diet: Public Service Media, Algorithms and the Multiple Dimensions of Diversity

Contact Info

Product

Resources

About