Ranking Pages by Topology and Popularity within Web Sites

We compare two link analysis ranking methods of web pages in a site. The first, called Site Rank, is an adaptation of PageRank to the granularity of a web site and the second, called Popularity Rank, is based on the frequencies of user clicks on the outlinks in a page that are captured by navigation sessions of users through the web site. We ran experiments on artificially created web sites of different sizes and on two real data sets, employing the relative entropy to compare the distributions of the two ranking methods. For the real data sets we also employ a nonparametric measure, called Spearman's footrule, which we use to compare the top-ten web pages ranked by the two methods. Our main result is that the distributions of the Popularity Rank and Site Rank are surprisingly close to each other, implying that the topology of a web site is very instrumental in guiding users through the site. Thus, in practice, the Site Rank provides a reasonable first order approximation of the aggregate behaviour of users within a web site given by the Popularity Rank.     

考虑用户活跃度和项目流行度的基于项目最近邻的协同过滤算法

项目相关性度量是基于项目最近邻的协同过滤算法的关键。已有的项目相关性度量方法在数据集稀疏或推荐低流行度产品时会面临较大挑战,因此提出一种考虑用户活跃度和项目流行度的基于项目最近邻的协同过滤算法。该算法在度量两个项目的相关性时,若有记录只对两个项目之一有评分,则利用该记录所对应的评分用户的活跃度和被评价项目的流行度进行相关性惩罚,从而提高数据稀疏环境下低流行度产品被推荐的概率。实验表明,所提算法在保证评分预测精度的情况下提升了推荐结果的多样性和新颖性。     

The design and evaluation of accessibility on web navigation

Web sites have been deployed to create and sustain business competitiveness in a trend of emerging Web technologies and growing e-commerce. One critical success factor of e-commerce is the ability to allow information to be retrieved from a Web site in an efficient and effective manner. Such ability, being determined by both the Web site structure and the Web page organization, can be measured in terms of accessibility and popularity of Web pages. The relationship between accessibility and popularity of web pages is dynamic in nature and can be analyzed to enhance a Web design. Having observed the lack of means to measure information retrieval of a Web site, this paper purports to introduce a guideline to evaluate Web page accessibility based on several structural-based accessibility models where an innovative accessibility–popularity (A–P) analysis is deployed to measure and, thereby, to modify a Web structure. Both push (i.e. demand driven) strategies and pull (i.e. design driven) strategies are incorporated into such guideline. Further, accessibility models are analyzed and compared in order to identify appropriate applications for each model. The paper is concluded by a summary of future directions of the accessibility models.     

A circular skip-cluster scheme to support video-on-demand services

The major drawback of the existing cluster placement scheme is the long response time caused by admission control if the number of clusters and the number of users are large. A circular skip-cluster placement scheme is proposed to reduce the size of the data buffer as well as the system response time. Furthermore, the popularity of each video is different in the real world. We propose a new popularity-based data allocation scheme to allocate data units within a cluster such that the corresponding data units of these popular videos are stored in those cylinders at one end of each cluster. Due to a higher spatial locality within these hot cylinders, some data units requested by the users are stored in the same cylinder such that one seek operation, one rotation, and one transfer operation are required to retrieve these data units. Therefore, the time required to retrieve data for these requests can be reduced, thus also reducing the system response time. Based on our results, the buffer size and the system response time can be reduced by half or more. These findings are essential for constructing video-on-demand systems that provide satisfactory performance.     

Looking at both sides of relationship dynamics in virtual communities: A social exchange theoretical lens

This study draws on social exchange theory to investigate the relationship dynamics of the member–community dyad in virtual community settings. Using a longitudinal design and multiple measurement sources, the findings indicate that social and task communication styles have unique effects on members’ community commitment velocity, which in turn influences member gratitude and entitlement behaviors. A moderated path analysis demonstrates that the effects of social and task communication styles on community commitment velocity are both positively moderated by attachment anxiety and attachment avoidance. It also reveals that members’ level of popularity augments the influence of community commitment velocity on member entitlement behavior.     

Interdisciplinary study on popularity prediction of social classified hot online events in China

We offer an interdisciplinary study of computer science and social science, analyzing behavior surrounding three types of online events: political events, social events, and non-public events. Based on the intrinsic characteristics of the three event types, this paper creates an effective method to predict such events. We continuously followed and recorded data every 10 min for 10 months from September 14, 2012 to July 11, 2013, and collected over 14 million “hot” posts from Sina Weibo, the largest microblogging provider in China. After removing spammers and noises, we developed a database of 4180 hot online events and 7,761,395 threads. We found that people’s online behavior regarding event types varies in terms of follow-up statistics and the predictability of events. The Chinese are, typically, quite concerned with social affairs that relate most closely to their personal interests and preferences. People tend to cluster around political events more often than social events and non-public events. This is demonstrated by an algorithm embedded with a clustering growth pattern of events, which predicts the popularity of online political events above others. The statistical findings are justified by Habermas’ public sphere theory and the theory of vertical/horizontal collectivism/individualism. This research provides an interesting piece of computational social science work to assist in the analysis of incentives concerning China’s collective events.     

午睡在中国的流行性及原因分析

午睡在我国到底有多么流行?午睡在我国流行的原因是什么?为了回答这两个问题,作者在国内进行了一项问卷调查。调查结果表明,午睡在我国的确非常流行,占被调查人80％以上,比国外进行过同类研究的国家都要流行。午睡的主观原因有2个,即人们认为午睡后下午和晚上的精神会好些,午睡可以弥补晚上睡眠的不足。午睡的客观原因是人们有时间午睡。绝大多数被调查者认为午睡是有必要或者非常必要。这些调查结果对如何在我国实行午睡的改革有重要参考价值。     

Using product popularity to stimulate choice for light products in supermarkets: An examination in virtual reality

With the use of virtual reality, this study investigates the potential for popularity cues to stimulate consumers to choose healthier products (versus regular products) within product categories in a supermarket context. Popularity may increase perceived product quality and perceived certainty about that quality. Healthier alternatives (i.e., light products) within a category often have a taste disadvantage compared to regular counterparts as a result of their trimmed down nutrient profiles (low in fat, sugar or salt). Because consumers are more uncertain about the taste of these light products, a popularity cue may have a larger effect on product choice for these options than on the choice for regular products. This hypothesis is tested in an experiment using virtual reality to realistically display a supermarket context. In the main experiment, a sample of 300 consumers was asked to shop for specific groceries, using existing brands and products. The results demonstrate that consumers are more likely to choose light products when these are combined with a popularity cue. In contrast, the popularity cue did not affect choice for the regular alternatives. These results have important implications for research, and for public-policy makers and food retailers that aim to stimulate healthful consumption.     

Generating request streams on Big Data using clustered renewal processes

The performance evaluation of large file systems, such as storage and media streaming, motivates scalable generation of representative traces. We focus on two key characteristics of traces, popularity and temporal locality. The common practice of using a system-wide distribution obscures per-object behavior, which is important for system evaluation. We propose a model based on delayed renewal processes which, by sampling interarrival times for each object, accurately reproduces popularity and temporal locality for the trace. A lightweight version reduces the dimension of the model with statistical clustering. It is workload-agnostic and object type-aware, suitable for testing emerging workloads and ‘what-if’ scenarios. We implemented a synthetic trace generator and validated it using: (1) a Big Data storage (HDFS) workload from Yahoo!, (2) a trace from a feature animation company, and (3) a streaming media workload. Two case studies in caching and replicated distributed storage systems show that our traces produce application-level results similar to the real workload. The trace generator is fast and readily scales to a system of 4.3 million files. It outperforms existing models in terms of accurately reproducing the characteristics of the real trace.     

基于深度学习的信息级联预测方法综述

在线社交媒体极大地促进了信息的产生和传递,加速了海量信息之间的传播与交互,使预测信息级联的重要性逐渐突显。近年来,深度学习已经被广泛用于信息级联预测(Information Cascade Prediction)领域。文中主要对基于深度学习的信息级联预测方法的研究现状与经典算法进行分类、梳理与总结。根据信息级联特征刻画的侧重点不同,将基于深度学习的信息级联预测方法分为时序信息级联预测方法与拓扑信息级联预测方法,并进一步将时序信息级联预测方法分为基于随机游走(Random Walk)的方法与基于扩散路径的方法,将拓扑信息级联预测方法分为基于全局拓扑结构的方法与基于邻域聚合的方法;并对每类方法进行详细的原理阐述与优缺点介绍,介绍了信息级联预测领域常用的数据集与评价指标,在宏观与微观两种信息级联预测场景下对基于深度学习的信息级联预测算法进行实验对比,并讨论了一些信息级联预测算法中常用的算法实现细节。最后,总结了该领域未来可能的研究方向与发展趋势。