A peek into the future

Ahmed, Mohamed; Spagna, Stella; Huici, Felipe; Niccolini, Saverio

doi:10.1145/2433396.2433473

Cited by 133 publications

(11 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The prediction is computed using a combination of popularity metrics, as well as content‐ and time‐related features. Finally, in Ahmed et al (), a clustering technique is proposed to predict the future popularity of web content using two simple features based on the past popularity of the content.…”

Section: Related Workmentioning

confidence: 99%

“…First, we examine the temporal features of online news consumption. Our choice is motivated by Ahmed et al () and Marujo et al (), where the authors successfully employ date and time information as features for their prediction tasks. As the correlation coefficient scores shown in Table suggest, this category of features is weakly correlated with the Tweets and Pageviews metrics.…”

Section: Featuresmentioning

confidence: 99%

“…There is also a fairly large number of studies on popularity prediction in the context of online news (Ahmed, Spagna, Huici, & Niccolini, ; Freyne, Berkovsky, Daly, & Geyer, ; Garimella & Castillo, ; Gupta, Gao, Zhai, & Han, ; Jamali & Rangwala, ; Lerman & Hogg, ; Marujo, Bugalho, da Silva Neto, Gershman, & Carbonell, ; Szabo & Huberman, ; Tatar et al, ; Tsagkias, Weerkamp, & de Rijke, ). Most of these works focused on prediction, exploiting early‐stage popularity metrics (i.e., not cold‐start prediction).…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

On the feasibility of predicting popular news at cold start

Arapakis

Cambazoğlu

Lalmas

2016

Asso for Info Science & Tech

View full text Add to dashboard Cite

Prominent news sites on the web provide hundreds of news articles daily. The abundance of news content competing to attract online attention, coupled with the manual effort involved in article selection, necessitates the timely prediction of future popularity of these news articles. The future popularity of a news article can be estimated using signals indicating the article's penetration in social media (e.g., number of tweets) in addition to traditional web analytics (e.g., number of page views). In practice, it is important to make such estimations as early as possible, preferably before the article is made available on the news site (i.e., at cold start). In this paper we perform a study on cold‐start news popularity prediction using a collection of 13,319 news articles obtained from Yahoo News, a major news provider. We characterize the popularity of news articles through a set of online metrics and try to predict their values across time using machine learning techniques on a large collection of features obtained from various sources. Our findings indicate that predicting news popularity at cold start is a difficult task, contrary to the findings of a prior work on the same topic. Most articles' popularity may not be accurately anticipated solely on the basis of content features, without having the early‐stage popularity values.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Featuresmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

On the feasibility of predicting popular news at cold start

Arapakis

Cambazoğlu

Lalmas

2016

Asso for Info Science & Tech

View full text Add to dashboard Cite

show abstract

“…In addition to user preferences, the timing of posting has been considered a crucial variable for describing user decisions to contribute or not (Kalman, Ravid, Raban, & Rafaeli, 2006; Lee & Lewis, 2012; Tsagkias, Weerkamp, & De Rijke, 2010). Previous studies have included the temporal dimension of user commenting to predict comment popularity (e.g., Ahmed, Spagna, Huici, & Niccolini, 2013; Kalman et al, 2006; Tatar et al, 2011). For example, Kalman and colleagues (2006) identified a binary temporal behavior in synchronous email exchanges—where users responded either extremely fast or relatively slow.…”

Section: Social Influence and The Cut Frameworkmentioning

confidence: 99%

“Information Warfare” and Online News Commenting: Analyzing Forces of Social Influence Through Location-Based Commenting User Typology

Zelenkauskaite¹,

Balduccini²

2017

Social Media + Society

View full text Add to dashboard Cite

While most of the online participation research assumes the Western notion of fulfilling deliberative practices, online contexts have been also found to be an active battleground for so-called information warfare. To test the potential for online comments being used by cross-national political opponents, we analyze the case of online comments on the most active Lithuanian online news portal’s Russian-language edition. This news portal presents itself as a unique case, since Russia was found to engage in a support or mobilization of the compatriot community in the Baltic States by exercising “soft influence.”To analyze such “soft influence” our study proposed a framework called Commenting User Typology (CUT). The CUT combines three types of data points: a) content level (topic of the story category), b) user level (frequency of posting), and c) the timing and location of the posting to contribute to studies of online social influence. We have applied the CUT framework to the commenting practices of 4,940 users who produced 34,038 comments over a month. In the analysis, we found that the majority of the users’ posts were low in topic variety and infrequent; followed by low in variety and frequent; high in topic variety and frequent; and high in topic variety and infrequent in posting. Moreover, while around a half of the comments come from Lithuania, 16% from Russia, yet more than 7% were found to mask IP addresses. Also, the results of this study suggest that more active participants tend to participate in more topic areas and that posting occurs faster in more active topic areas. The implications of this study relate to online news portals overall as well as online news portals in the non-Western contexts or electoral totalitarian regimes as Russia. We argue that user-based analysis, such as the CUT-based approach we proposed, can be particularly relevant for contexts where deliberative practices are still in negotiation.

show abstract

“…The works by [Myers and Leskovec 2014;Naaman et al 2011] discuss factors that affect topic trends and the bursty dynamics in Twitter, and hashtags in microposts are utilized by [Tsur and Rappoport 2012] for predicting topic propagation. Regression and classification algorithms are used by [Asur et al 2011;Bandari et al 2012] to predict news popularity in social media, temporal patterns evolution and state transition based topic popularity prediction methods are discussed by [Ahmed et al 2013], and Gradient Boosted Decision Tree model for microposts show counts is proposed by [Kupavskii et al 2013]. There are also other purposes of topic analysis in social networks.…”

Section: Related Workmentioning

confidence: 99%

Cost-Effective Online Trending Topic Detection and Popularity Prediction in Microblogging

Miao

Chen

Fang

et al. 2016

ACM Trans. Inf. Syst.

View full text Add to dashboard Cite

Identifying topic trends on microblogging services such as Twitter and estimating those topics' future popularity have great academic and business value, especially when the operations can be done in real time. For any third party, however, capturing and processing such huge volumes of real-time data in microblogs are almost infeasible tasks, as there always exist API request limits, monitoring and computing budgets, as well as timeliness requirements. To deal with these challenges, we propose a cost-effective system framework with algorithms that can automatically select a subset of representative users in microblogging networks in offline, under given cost constraints. Then the proposed system can online monitor and utilize only these selected users' real-time microposts to detect the overall trending topics and predict their future popularity among the whole microblogging network. Therefore, our proposed system framework is practical for realtime usage as it avoids the high cost in capturing and processing full real-time data, while not compromising detection and prediction performance under given cost constraints. Experiments with real microblogs dataset show that by tracking only 500 users out of 0.6 million users and processing no more than 30,000 microposts daily, about 92% trending topics could be detected and predicted by the proposed system, and on average more than 10 hours earlier than they appear in official trends list.

show abstract

A peek into the future

Cited by 133 publications

References 28 publications

On the feasibility of predicting popular news at cold start

On the feasibility of predicting popular news at cold start

“Information Warfare” and Online News Commenting: Analyzing Forces of Social Influence Through Location-Based Commenting User Typology

Cost-Effective Online Trending Topic Detection and Popularity Prediction in Microblogging

Contact Info

Product

Resources

About