Online news recommendations based on topic modeling and online interest adjustment

2019 ◽  
Vol 119 (8) ◽  
pp. 1802-1818
Author(s):  
Duen-Ren Liu ◽  
Yu-Shan Liao ◽  
Jun-Yi Lu
Keyword(s):  
Topic Modeling ◽  
Recommendation System ◽  
Online News ◽  
User Preferences ◽  
Online Media ◽  
Content Type ◽  
Online Evaluation ◽  
Important Trend ◽  
News Recommendation ◽  
Click Through Rate

Purpose Providing online news recommendations to users has become an important trend for online media platforms, enabling them to attract more users. The purpose of this paper is to propose an online news recommendation system for recommending news articles to users when browsing news on online media platforms. Design/methodology/approach A Collaborative Semantic Topic Modeling (CSTM) method and an ensemble model (EM) are proposed to predict user preferences based on the combination of matrix factorization with articles’ semantic latent topics derived from word embedding and latent topic modeling. The proposed EM further integrates an online interest adjustment (OIA) mechanism to adjust users’ online recommendation lists based on their current news browsing. Findings This study evaluated the proposed approach using offline experiments, as well as an online evaluation on an existing online media platform. The evaluation shows that the proposed method can improve the recommendation quality and achieve better performance than other recommendation methods can. The online evaluation also shows that integrating the proposed method with OIA can improve the click-through rate for online news recommendation. Originality/value The novel CSTM and EM combined with OIA are proposed for news recommendation. The proposed novel recommendation system can improve the click-through rate of online news recommendations, thus increasing online media platforms’ commercial value.

Integrating collaborative topic modeling and diversity for movie recommendations during news browsing

Kybernetes ◽  
2019 ◽  
Vol 49 (11) ◽  
pp. 2633-2649
Author(s):  
Duen-Ren Liu ◽  
Yun-Cheng Chou ◽  
Ciao-Ting Jian
Keyword(s):  
Topic Modeling ◽  
Latent Dirichlet Allocation ◽  
Online News ◽  
Content Type ◽  
Online Evaluation ◽  
News Websites ◽  
News Website ◽  
Recommendation Diversity ◽  
Cross Type ◽  
News Online

Purpose Online news websites provide diverse article topics, such as fashion news, entertainment and movie information, to attract more users and create more benefits. Recommending movie information to users reading news online can enhance the impression of diverse information and may consequently improve benefits. Accordingly, providing online movie recommendations can improve users’ satisfactions with the website, and thus is an important trend for online news websites. This study aims to propose a novel online recommendation method for recommending movie information to users when they are browsing news articles. Design/methodology/approach Association rule mining is applied to users’ news and movie browsing to find latent associations between news and movies. A novel online recommendation approach is proposed based on latent Dirichlet allocation (LDA), enhanced collaborative topic modeling (ECTM) and the diversity of recommendations. The performance of proposed approach is evaluated via an online evaluation on a real news website. Findings The online evaluation results show that the click-through rate can be improved by the proposed hybrid method integrating recommendation diversity, LDA, ECTM and users’ online interests, which are adapted to the current browsing news. The experiment results also show that considering recommendation diversity can achieve better performance. Originality/value Existing studies had not investigated the problem of recommending movie information to users while they are reading news online. To address this problem, a novel hybrid recommendation method is proposed for dealing with cross-type recommendation tasks and the cold-start issue. Moreover, the proposed method is implemented and evaluated online in a real world news website, while such online evaluation is rarely conducted in related research. This work contributes to deriving user’s online preferences for cross-type recommendations by integrating recommendation diversity, LDA, ECTM and adaptive online interests. The research findings also contribute to increasing the commercial value of the online news websites.

scholarly journals Hot News Recommendation System from Heterogeneous Websites Based on Bayesian Model

2014 ◽  
Vol 2014 ◽  
pp. 1-8 ◽  
Author(s):  
Zhengyou Xia ◽  
Shengwu Xu ◽  
Ningzhong Liu ◽  
Zhengkang Zhao
Keyword(s):  
Bayesian Model ◽  
Recommendation System ◽  
Research Work ◽  
Joint Probability ◽  
Real Data ◽  
Online News ◽  
City Government ◽  
Data Sets ◽  
News Websites ◽  
News Recommendation

The most current news recommendations are suitable for news which comes from a single news website, not for news from different heterogeneous news websites. Previous researches about news recommender systems based on different strategies have been proposed to provide news personalization services for online news readers. However, little research work has been reported on utilizing hundreds of heterogeneous news websites to provide top hot news services for group customers (e.g., government staffs). In this paper, we propose a hot news recommendation model based on Bayesian model, which is from hundreds of different news websites. In the model, we determine whether the news is hot news by calculating the joint probability of the news. We evaluate and compare our proposed recommendation model with the results of human experts on the real data sets. Experimental results demonstrate the reliability and effectiveness of our method. We also implement this model in hot news recommendation system of Hangzhou city government in year 2013, which achieves very good results.

scholarly journals Incorporating popularity in a personalized news recommender system

2016 ◽  
Vol 2 ◽  
pp. e63 ◽  
Author(s):  
Nirmal Jonnalagedda ◽  
Susan Gauch ◽  
Kevin Labille ◽  
Sultan Alfarhood
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
Recommendation System ◽  
User Profile ◽  
Online News ◽  
User Interests ◽  
News Sources ◽  
Enormous Amount ◽  
News Recommendation

Online news reading has become a widely popular way to read news articles from news sources around the globe. With the enormous amount of news articles available, users are easily overwhelmed by information of little interest to them. News recommender systems help users manage this flood by recommending articles based on user interests rather than presenting articles in order of their occurrence. We present our research on developing personalized news recommendation system with the help of a popular micro-blogging service, “Twitter.” News articles are ranked based on the popularity of the article identified from Twitter’s public timeline. In addition, users construct profiles based on their interests and news articles are also ranked based on their match to the user profile. By integrating these two approaches, we present a hybrid news recommendation model that recommends interesting news articles to the user based on their popularity as well as their relevance to the user profile.

Using latent features for building an interpretable recommendation system

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ziming Zeng ◽  
Yu Shi ◽  
Lavinia Florentina Pieptea ◽  
Junhua Ding
Keyword(s):  
Recommendation System ◽  
User Preferences ◽  
User Preference ◽  
Data Sets ◽  
Historical Records ◽  
Data Link ◽  
Content Type ◽  
Fine Grained ◽  
Public Data ◽  
Latent Features

Purpose Aspects extracted from the user’s historical records are widely used to define user’s fine-grained preferences for building interpretable recommendation systems. As the aspects were extracted from the historical records, the aspects that represent user’s negative preferences cannot be identified because of their absence from the records. However, these latent aspects are also as important as those aspects representing user’s positive preferences for building a recommendation system. This paper aims to identify the user’s positive preferences and negative preferences for building an interpretable recommendation. Design/methodology/approach First, high-frequency tags are selected as aspects to describe user preferences in aspect-level. Second, user positive and negative preferences are calculated according to the positive and negative preference model, and the interaction between similar aspects is adopted to address the aspect sparsity problem. Finally, an experiment is designed to evaluate the effectiveness of the model. The code and the experiment data link is: https://github.com/shiyu108/Recommendation-system Findings Experimental results show the proposed approach outperformed the state-of-the-art methods in widely used public data sets. These latent aspects are also as important as those aspects representing the user’s positive preferences for building a recommendation system. Originality/value This paper provides a new approach that identifies and uses not only users’ positive preferences but also negative preferences, which can capture user preference precisely. Besides, the proposed model provides good interpretability.

Design and evaluation of a movie recommendation system showing a review for evoking interested

2017 ◽  
Vol 13 (1) ◽  
pp. 72-84 ◽  
Author(s):  
Yuto Ishida ◽  
Takahiro Uchiya ◽  
Ichi Takumi
Keyword(s):  
Recommendation System ◽  
User Profile ◽  
User Preferences ◽  
User Preference ◽  
Item Selection ◽  
Simple Method ◽  
Content Type ◽  
Goods And Services ◽  
Outcomes Evaluation ◽  
Movie Recommendation

Purpose In recent years, e-commerce (EC) sites dealing in various goods and services have increased along with internet popularity. Now, very few EC recommendation systems present a concrete reason for their recommendations. Therefore, because user preferences strongly influence outcomes, evaluation and selection are difficult for items, such as books, movies and luxury goods. The purpose of this paper is evoking interest by showing the review as a reason for a user’s decision-making factor. This paper aims to presents the development and introduction of a recommendation system that presents a review adapted to user preference. Design/methodology/approach The system presents a review to the user, which indicates the reason for matching the item contents and user preferences. Thereby, this system enables the creation of personalized reasons for recommendations. Findings Recommendation sentences conforming to user preferences are effective for item selection. Even with a simple method, in this paper, it was possible to present a review which is an item selection factor sufficient for the user. Originality/value This system can show a recommendation sentence that conforms to a user’s preferences merely from a user profile with the tag data of a product. This paper dealt in movies, but it can easily be applied even for other items.

Understanding the motivational values for the usage of specific online news media and users’ perception of information credibility

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Tunde Simeon Amosun ◽  
Chu Jianxun ◽  
Olayemi Hafeez Rufai ◽  
Muhideen Sayibu ◽  
Riffat Shahani ◽  
...  
Keyword(s):  
News Media ◽  
Structural Equation ◽  
Positive Impact ◽  
Online News ◽  
Equation Modeling ◽  
Online Media ◽  
Research Model ◽  
Content Type ◽  
Information Credibility ◽  
Motivational Values

Purpose The purpose of this study is to investigate the utilitarian value (UV), hedonic value (HV) and social value (SV) that make people use a certain type of online media website and how the usage of specific online media website impact the way people perceive online information credibility (OIC). A research model was also proposed to explain the essence of this study. Design/methodology/approach This study adopted the survey research methodology to empirically test the research model with 873 research participants from the University of Science and Technology of China and Anhui Medical University. Findings Results from structural equation modeling showed that UV and HV have a significant positive impact on the usage of print news media website (PNMW), usage of broadcast news media website (BNMW) and usage of social networking website (SNW). The SV was also found to have a significant positive impact on the usage of SNWs. The result also indicated that the usage of the PNMW and the usage of the BNMW by online users have a significantly positive impact on high rating of OIC. However, the result showed that the usage of SNW does not have a significant positive impact on the high rating of OIC. Originality/value Findings in this study provided substantial contributions toward the advancement of the uses and gratification theoretical framework by unraveling how certain motivational values can influence online media users’ preferences for specific online media websites, as well as showing how specific online media websites affect online users’ perception of OIC.

Constructing linkage between libraries and up-to-date news

2016 ◽  
Vol 34 (2) ◽  
pp. 301-313
Author(s):  
Wei Yu ◽  
Junpeng Chen
Keyword(s):  
Matrix Factorization ◽  
Recommendation System ◽  
Online News ◽  
Daily News ◽  
Content Type ◽  
Human Evaluation ◽  
Weighted Matrix ◽  
Effective Performance ◽  
Novel Method ◽  
Practical Implications

Purpose – The purpose of this paper is to construct the linkage between libraries and up-to-date news. This study developed a system to recommend libraries’ resources to those daily news readers who are interested in the topics of the target news. The analysis of experiments results served as the reference for the development and improvement of linking libraries’ resources with other web resources. Design/methodology/approach – Up-to-date news were gathered through the news feeds to make the integration with the libraries’ records. In task 1, the libraries’ records were linked and recommended to the target libraries’ records which are of the same topics. In task 2, the system aimed to find the relevant libraries’ records for target news. Three recommendation methods were compared in both tasks to find the most effective approach to the system. Findings – Experiment results showed that: at first, in task 1, the system can assign the libraries’ records of the related topics effectively; second, in task 2, the recommending system can obtain a satisfied recall hit rate through human evaluation. Therefore, regarding the popularity of the daily news online, the linkage and recommendation with the libraries’ resources can increase the visibility of the libraries’ resources and eventually promote the information consuming in libraries. Practical implications – The authors have confirmed, using three matrix factorization methods, that weighted matrix factorization used in the libraries’ records recommendation system, could achieve better performance than the other two. Based on the research, the libraries could incorporate the online news and libraries’ resources in practice. Originality/value – To increase the visibility and promote information consuming of libraries, this study proposed a novel method to construct the linkage between library and up-to-date news. The results of data analysis indicate that recommendation of libraries resources through the daily news can achieve effective performance. Thus, it can be inferred that the research results of this study are representative and have practical values in real world practice.

scholarly journals Topic Modeling in the News Document on Sustainable Development Goals

2021 ◽  
Vol 5 (3) ◽  
pp. 82
Author(s):  
Hidayatul Fitri ◽  
Widyawan Widyawan ◽  
Indah Soesanti
Keyword(s):  
Sustainable Development ◽  
Topic Modeling ◽  
Sustainable Development Goals ◽  
Latent Dirichlet Allocation ◽  
Online News ◽  
Online Media ◽  
Decent Work ◽  
Development Goals ◽  
The Government ◽  
The Right

Indonesia is a developing country and supports the program of the Sustainable Development Goals (SDGs) which consist of 17 goals. SDGs is not only the government’s duty, but a shared duty from any elements. Online media has a crucial role in implementing goals of Indonesia’s SDG. Information published in online news related to the SDGs is an important consideration for the government, society, and all elements. Categorizing news manually to find out news topics is very time-consuming and done by the ability of news editors. News presented by online media on the news site can be used as topic modeling, where hidden topics can be found in the news on online media. Topic modeling will classify data based on a particular topic and determine the relationship between text. Latent Dirichlet allocation (LDA) is one of the methods on topic modeling to find out the trend of topics of SDGs news. Based on the result of this research, the implementation of LDA is the right choice for finding topics in a document. The result of topic modeling with k = 17 obtained the highest coherence score of 0.5405 on topic 8. Topic 8 discussed news related to the eighth SDGs goals, namely decent work and economic growth. This categorization was based on words formed after the LDA process. Then, topic 5 discussed the news on the 17th SDGs goals, namely partnerships for the goals. Topic 6 discussed the news of the first SDGs, namely no poverty.

New Indian online rules risk state overreach and abuse

2021 ◽  
Keyword(s):  
Social Media ◽  
News Media ◽  
Public Discourse ◽  
Hate Speech ◽  
Central Government ◽  
Service Providers ◽  
Online News ◽  
Content Type ◽  
Digital Service ◽  
The Government

Significance The new rules follow a stand-off between Twitter and the central government last month over some posts and accounts. The government has used this stand-off as an opportunity not only to tighten rules governing social media, including Twitter, WhatsApp, Facebook and LinkedIn, but also those for other digital service providers including news publishers and entertainment streaming companies. Impacts Government moves against dominant social media platforms will boost the appeal of smaller platforms with light or no content moderation. Hate speech and harmful disinformation are especially hard to control and curb on smaller platforms. The new rules will have a chilling effect on online public discourse, increasing self-censorship (at the very least). Government action against online news media would undercut fundamental democratic freedoms and the right to dissent. Since US-based companies dominate key segments of the Indian digital market, India’s restrictive rules could mar India-US ties.

Trust mechanisms underlying the self-efficacy-rumour use relationship

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Ping Wang ◽  
Yixia Hu ◽  
Qiao Li ◽  
Hanqin Yang
Keyword(s):  
Social Media ◽  
Self Efficacy ◽  
Structural Equation ◽  
Online News ◽  
Dual Process ◽  
Content Type ◽  
Consumptive Use ◽  
Negative Effect ◽  
Journalism Students ◽  
Antecedents And Consequences

Purpose Journalism students, a special user group with the dual perspective of both social media general users and online journalists, and their trust in rumours is a valued but understudied topic in relation to preparing rational information users and professionals for rumour control. To reveal these trust mechanisms, this paper aims to identify salient psychological and behavioural factors related to journalism students’ different levels of trust. Design/methodology/approach Using structural equation modelling to analyse the survey data of 234 journalism students, this paper tested a theoretical model that considers self-efficacy and the expressive and consumptive use of social media rumours as the antecedents and consequences of trust belief and trust action, respectively. Findings Self-efficacy has a positive effect on trust belief but a negative effect on trust action. Trust belief positively affects expressive use of rumours, whereas trust action negatively affects consumptive use. Practical implications This study contributes to the cultivation of future online news gatekeepers. Originality/value This paper distinguishes journalism students’ trust mechanisms from those of general users and online journalists. The integration of dual process theories provides insights into trust-building processes related to rumours and advances the understanding of the anchoring and adjustment effects of self-efficacy on trust.

Sign in / Sign up

Export Citation Format

Share Document