Understanding how users behave when they connect to social networking sites creates opportunities for better interface design, richer studies of interactions, and improved design content distribution systems. In this paper, we present a first kind analysis user workloads in online networks. Our study is based on detailed clickstream data, collected over 12-day period, summarizing HTTP sessions 37,024 who accessed four popular networks: Orkut, MySpace, Hi5, LinkedIn. The data were from...

The authors propose models for both temporal and spatial locality of reference in streams requests arriving at Web servers. They show that simple based on document popularity alone are insufficient capturing either or locality. Instead, they rely an equivalent, but numerical, representation a stream: stack distance trace. can be characterized by the marginal distribution trace, typical distributions compare their cache performance to traces. also stream using notion self-similarity....

Non-profits, as well the media, have hypothesized existence of a radicalization pipeline on YouTube, claiming that users systematically progress towards more extreme content platform. Yet, there is to date no substantial quantitative evidence this alleged pipeline. To close gap, we conduct large-scale audit user YouTube. We analyze 330,925 videos posted 349 channels, which broadly classified into four types: Media, Alt-lite, Intellectual Dark Web (I.D.W.), and Alt-right. According...

AI-based systems are "black boxes," resulting in massive information asymmetries between the developers of such and consumers policymakers. In order to bridge this gap, article proposes a conceptual framework for thinking about governance AI.

Twitter is a unique social media channel, in the sense that users discuss and talk about most diverse topics, including their health conditions. In this paper we analyze how Dengue epidemic reflected on to what extent information can be used for sake of surveillance. mosquito-borne infectious disease leading cause illness death tropical subtropical regions, Brazil. We propose an active surveillance methodology based four dimensions: volume, location, time public perception. First explore...

With the growing popularity and usage of online social media services, people now have accounts (some times several) on multiple diverse services like Facebook, Linked In, Twitter You Tube. Publicly available information can be used to create a digital footprint any user using these services. Generating such footprints very useful for personalization, profile management, detecting malicious behavior users. A important application analyzing users' is protect users from potential privacy...

Current approaches to characterize and detect hate speech focus on content posted in Online Social Networks (OSNs). They face shortcomings get the full picture of due its subjectivity noisiness OSN text. This work partially addresses these issues by shifting towards users. We obtain a sample Twitter's retweet graph with 100,386 users annotate 4,972 as hateful or normal, also find 668 suspended after 4 months. Our analysis shows that hateful/suspended differ from normal/active ones terms...

Article Free Access Share on A methodology for workload characterization of E-commerce sites Authors: Daniel A. Menascé Dept. Computer Science, George Mason University, Fairfax, VA VAView Profile , Virgilio F. Almeida Univ. Federal de Minas Gerais, Belo Horizonte, MG 30161, Brazil BrazilView Rodrigo Fonseca Marco Mendes Authors Info & Claims EC '99: Proceedings the 1st ACM conference Electronic commerceNovember 1999 Pages 119–128https://doi.org/10.1145/336992.337024Published:01 November...

A number of online video social networks, out which YouTube is the most popular, provides features that allow users to post a as response discussion topic. These open opportunities for introduce polluted content, or simply pollution, into system. For instance, spammers may an unrelated popular one aiming at increasing likelihood being viewed by larger users. Moreover, opportunistic users--promoters--may try gain visibility specific posting large (potentially unrelated) responses boost rank...

Online social networks pose an interesting problem: how to best characterize the different classes of user behavior. Traditionally, behavior characterization methods, based on individual features, are not appropriate for online networking sites. In these environments, users interact with site and other through a series multiple interfaces that let them upload view content, choose friends, rank favorite subscribe do many interactions. Different interaction patterns can be observed groups...

Traditionally, users have discovered information on the Web by browsing or searching. Recently, word-of-mouth has emerged as a popular way of discovering Web, particularly social networking sites like Facebook and Twitter. On these sites, discover content following URLs posted their friends. Such based discovery become major driver traffic to many today. To better understand this phenomenon, in paper we present detailed analysis exchange among Twitter users. Among our key findings, show that...

Real-time interaction, which enables live discussions, has become a key feature of most Web applications. In such an environment, the ability to automatically analyze user opinions and sentiments as discussions develop is powerful resource known real time sentiment analysis. However, this task comes with several challenges, including need deal highly dynamic textual content that characterized by changes in vocabulary its subjective meaning lack labeled data needed support supervised...

Online social networks (OSNs) have become popular platforms for people to connect and interact with each other. Among those networks, Pinterest has recently noteworthy its growth promotion of visual over textual content. The purpose this study is analyze image-based network in a gender-sensitive fashion, order un- derstand (i) user motivation usage pattern the network, (ii) how communications interactions happen (iii) users describe themselves others. This work based on more than 220 million...

We present a thorough characterization of what we believe to be the first significant live Internet streaming media workload in scientific literature. Our over 3.5 million requests spanning 28-day period is done at three increasingly granular levels, corresponding clients, sessions, and transfers. findings support two important conclusions. First, show that nature interactions between users objects fundamentally different for versus stored objects. Access user driven, whereas access object...

We present what we believe to be the first thorough characterization of live streaming media content delivered over Internet. Our 3.5 million requests spanning a 28-day period is done at three increasingly granular levels, corresponding clients, sessions, and transfers. findings support two important conclusions. First, show that nature interactions between users objects fundamentally different for versus stored objects. Access user driven, whereas access object driven. This reversal...

Article Free Access Share on In search of invariants for e-business workloads Authors: Daniel Menascé Dept. Computer Science, George Mason University, Fairfax, VA VAView Profile , Virgílio Almeida Univ. Fed. Minas Gerais, Belo Horizonte, MG 31270, Brazil BrazilView Rudolf Riedi Electrical and Comp. Engineering, Rice Houston, TX TXView Flávia Ribeiro Rodrigo Fonseca Wagner Meira Authors Info & Claims EC '00: Proceedings the 2nd ACM conference Electronic commerceOctober 2000 Pages...