Predicting the Popularity of Online Content

Szabó, Gábor; Huberman, Bernardo A.

doi:10.2139/ssrn.1295610

Cited by 111 publications

(158 citation statements)

References 22 publications

(11 reference statements)

Supporting

Mentioning

154

Contrasting

Unclassified

Order By: Relevance

“…Here, we used a news life-cycle as the content model as it is a striking instance of dynamic content models. Since many newsworthy events are generated in a day [11]- [13] and instantaneous reports are important for news articles [14], [15], the frequency of generation of news content is higher and the change in the popularity of content is much more frequent those for entertainment.…”

Section: News Content Modelmentioning

confidence: 99%

See 1 more Smart Citation

An Approximated Selection Algorithm for Combinations of Content with Virtual Local Server for Traffic Localization in Peer-Assisted Content Delivery Networks

Maki

Shinkuma

Takahashi

2013

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Naoya MAKI †a) , Student Member, Ryoichi SHINKUMA †b) , Senior Member, and Tatsuro TAKAHASHI †c) , Fellow SUMMARYOur prior papers proposed a traffic engineering scheme to further localize traffic in peer-assisted content delivery networks (CDNs). This scheme periodically combines the content files and allows them to obtain the combined content files while keeping the price unchanged from the single-content price in order to induce altruistic clients to download content files that are most likely to contribute to localizing network traffic. However, the selection algorithm in our prior work determined which and when content files should be combined according to the cache states of all clients, which is a kind of unrealistic assumption in terms of computational complexity. This paper proposes a new concept of virtual local server to reduce the computational complexity. We could say that the source server in our mechanism has a virtual caching network inside that reflects the cache states of all clients in the 'actual' caching network and combines content files based on the virtual caching network. In this paper, without determining virtual caching network according to the cache states of all clients, we approximately estimated the virtual caching network from the cache states of the virtual local server of the local domain, which is the aggregated cache state of only altruistic clients in a local domain. Furthermore, we proposed a content selection algorithm based on a virtual caching network. In this paper, we used news life-cycle model as a content model that had the severe changes in cache states, which was a striking instance of dynamic content models. Computer simulations confirmed that our proposed algorithm successfully localized network traffic.

show abstract

Section: News Content Modelmentioning

confidence: 99%

“…Therefore, the rate of decline in the request probability of news content is sharper than that in entertainment content and does not follow Zipf's law. Some conventional research has analyzed and reported the life cycle of news articles [14]- [18]. Most of them can be approximated by monotonic decrease models.…”

Section: Request Probabilitymentioning

confidence: 99%

An Approximated Selection Algorithm for Combinations of Content with Virtual Local Server for Traffic Localization in Peer-Assisted Content Delivery Networks

Maki

Shinkuma

Takahashi

2013

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Also, our work is focused on generic events unlike movie forecasts (Asur and Huberman, 2010) where specific features like Hollywood Stock Exchange time series could be exploited as features. Besides Twitter, there has been work on future popularity of social media content on Digg and Youtube (Lerman and Hogg, 2010;Szabo and Huberman, 2010). But, the design of the user interface of Twitter and Digg are quite different resulting in a huge difference in social dynamics of the two networks.…”

Section: Predictive Analysis On Twitter and Other Platformsmentioning

confidence: 99%

Predicting future popularity trend of events in microblogging platforms

Gupta¹,

Gao

Zhai³

et al. 2012

Proc of Assoc for Info

View full text Add to dashboard Cite

The fast information sharing on Twitter from millions of users all over the world leads to almost real-time reporting of events. It is extremely important for business and administrative decision makers to learn events' popularity as quickly as possible, as it can buy extra precious time for them to make informed decisions. Therefore, we introduce the problem of predicting future popularity trend of events on microblogging platforms. Traditionally, trend prediction has been performed by using time series analysis of past popularity to forecast the future popularity changes. As we can encode the rich Twitter dynamics using a rich variety of features from microblogging data, we explore regression, classification and hybrid approaches, using a large set of popularity, social and event features, to predict event popularity. Experimental results on two real datasets of 18382 events extracted from ~133 million tweets show the effectiveness of the extracted features and learning approaches. The predicted popularity trend of events can be directly used for a variety of applications including recommendation systems, ad keywords bidding price decisions, stock trading decisions, dynamic ticket pricing for sports events, etc.

show abstract

“…Using analytical methods from social networks to investigate the macroscopic characteristics of the network of videos, works have studied the impact of user generated content in underlying video-on-demand architectures [6,7], and in local networks [9,23]. By concentrating specifically on videos, research has also studied the daily cycles of video reception (based on the number of views), in an attempt to automatically predict their popularity [22]. In relation to video content, several studies have manually coded samples of videos to categorize the types and properties of YouTube content to gain understanding about user-generated media production [13,12,5].…”

Section: Research In Youtubementioning

confidence: 99%

Call Me Guru: User Categories and Large-Scale Behavior in YouTube

Biel

Gática-Pérez

2011

Social Media Modeling and Computing

View full text Add to dashboard Cite

While existing studies on YouTube's massive user-generated video content have mostly focused on the analysis of videos, their characteristics, and network properties, little attention has been paid to the analysis of users' long-term behavior as it relates to the roles they self-define and (explicitly or not) play in the site. In this chapter, we present a statistical analysis of aggregated user behavior in YouTube from the perspective of user categories, a feature that allows people to ascribe to popular roles and to potentially reach certain communities. Using a sample of 270,000 users, we found that a high level of interaction and participation is concentrated on a relatively small, yet significant, group of users, following recognizable patterns of personal and social involvement. Based on our analysis, we also show that by using simple behavioral features from user profiles, people can be automatically classified according to their category with accuracy rates of up to 73%.

show abstract

Predicting the Popularity of Online Content

Cited by 111 publications

References 22 publications

An Approximated Selection Algorithm for Combinations of Content with Virtual Local Server for Traffic Localization in Peer-Assisted Content Delivery Networks

An Approximated Selection Algorithm for Combinations of Content with Virtual Local Server for Traffic Localization in Peer-Assisted Content Delivery Networks

Predicting future popularity trend of events in microblogging platforms

Call Me Guru: User Categories and Large-Scale Behavior in YouTube

Contact Info

Product

Resources

About