Topology-Based Recommendation of Users in Micro-Blogging Communities

Nowadays,more and more users share real-time news and information in micro-blogging communities such as Twitter,Tumblr or Plurk.In these sites,information is shared via a followers/followees social network structure in which a follower will receive all the micro-blogs from the users he/she follows,named followees.With the increasing number of registered users in this kind of sites,finding relevant and reliable sources of information becomes essential.The reduced number of characters present in micro-posts along with the informal language commonly used in these sites make it difficult to apply standard content-based approaches to the problem of user recommendation.To address this problem,we propose an algorithm for recommending relevant users that explores the topology of the network considering different factors that allow us to identify users that can be considered good information sources.Experimental evaluation conducted with a group of users is reported,demonstrating the potential of the approach.     

Supporting Creativity through the Interactive Exploratory Search Paradigm

In creative tasks, there is a need to explore the space of available information in order to come up with diverse views before converging to a solution. In such tasks, typical search engines that follow the direct search paradigm fail to inspire users. It is hypothesized that contrary to typical engines, interactive exploratory search, which aims at revealing latent, alternative directions in the information space enabling user orientation and engagement, is better suited to assist users in their quest for serendipitous discoveries and inspiration. In this study, an interactive exploratory search tool that combines diversification of content and sources with a user interface design that visualizes clues from the social chatter – generated with micro-blogging services such as Twitter – and lets users interactively explore the available information space is presented. A profiling service and recommendation module in charge of delivering personalized social content complements the setting. A pilot and two task-based user studies comparing our system to a query-based baseline indicate that our system significantly improves inspirational discoveries by providing access to more interesting and serendipitous information.     

Discerning individual interests and shared interests for social user profiling

Traditionally, research about social user profiling assumes that users share some similar interests with their followees. However, it lacks the studies on what topic and to what extent their interests are similar. Our study in online sharing sites reveals that besides shared interests between followers and followees, users do maintain some individual interests which differ from their followees. Thus, for better social user profiling we need to discern individual interests (capturing the uniqueness of users) and shared interests (capturing the commonality of neighboring users) of the users in the connected world. To achieve this, we extend the matrix factorization model by incorporating both individual and shared interests, and also learn the multi-faceted similarities unsupervisedly. The proposed method can be applied to many applications, such as rating prediction, item level social influence maximization and so on. Experimental results on real-world datasets show that our work can be applied to improve the performance of social rating. Also, it can reveal some interesting findings, such as who likes the “controversial” items most, and who is the most influential in attracting their followers to rate an item.     

Detecting and tagging users’ social circles in social media

As a media and communication platform, microblog becomes more popular around the world. Most users follow a large number of celebrities and public medias on microblog; however, these celebrities do not necessarily follow all their fans. Such one-way relationship abounds in ego network and is displayed by the forms of users’ followees and followers, which make it difficult to identify users’ real friends who are contained in merged lists of followees and followers. The aim of this paper is to propose a general algorithm for detecting users’ real friends in social media and dividing them into different social circles automatically according to the closeness of their relationships. Then we analyze these social circles and detect social attributes of these social circles. To verify the effectiveness of the proposed algorithm, we build a microblog application which displays algorithm results of social circles for users and enables users to adjust proposed results according to her/his real social circles. We demonstrate that our algorithm is superior to the traditional clustering method in terms of F value and mean average precision. Furthermore, our method of tagging social attributes of social circles gets high performance by NDCG (normalized discounted cumulative gain).     

基于社交关系的微博主题情感挖掘

微博情感分析是社交媒体挖掘中的重要任务之一,在个性化推荐、舆情分析等方面具有重要的理论和应用价值.挖掘性能良好且可同步进行文档主题分析与情感分析的主题情感模型近来在以微博为代表的社交媒体情感分析中备受关注。然而,绝大多数现有主题情感模型都简单地假设不同微博的情感极性是互相独立,这与微博生态的现实状况不相一致的,从而导致这些模型无法对用户的真实情感进行有效建模。基于此,本文综合考虑了微博用户相互关联的事实,提出基于LDA和微博用户关系的主题情感模型SRTSM,该模型在LDA中加入情感层与微博用户关系参数,利用微博用户关系与微博主题学习微博的情感极性。新浪微博真实数据集上的大量实验表明,与代表性算法JST、Sentiment-LDA与DPLDA相比较,SRTSM模型能对用户真实情感与讨论主题进行更加有效的分析建模.     

基于BLSTM-Attention神经网络模型的化工事故分类

化工事故新闻数据包含新闻内容,标题以及新闻来源等方面信息,新闻内容的文本对上下文具有较强的依赖性.为了更准确地提取文本特征并提高化工事故分类的准确性,该文提出了一种基于Attention机制的双向LSTM （BLSTM-Attention）神经网络模型对化工新闻文本进行特征提取并实现文本分类.BLSTM-Attention神经网络模型能够结合文本上下文语义信息,通过正向和反向的角度来提取事故新闻的文本特征;考虑到事故新闻中不同词对文本的贡献不大相同,加入Attention机制对不同词和句子分配不同权重.最后,将该文提出的分类方法与Naive-Bayes、CNN、RNN、BLSTM分类方法在相同的化工事故新闻数据集上进行实验对比.实验结果表明：该文提出的神经网络模型BLSTM-Attention神在化工数据集上的效果更优于其他分类方法模型.     

基于信息内容和拓扑关系的社会媒体用户兴趣分类

随着社会的发展,信息已经成为社会发展越来越重要的部分,人类的信息传播活动越来越明显地展示出分众特征,对用户的分类成为人类信息活动的一个重要研究课题.从这一目标出发,分别基于信息内容、拓扑关系和两者综合的方法,按兴趣主题对社会媒体用户进行分类.对于基于信息内容的用户分类,采用LDA主题模型从用户所发布的内容中提取其主题分布,基于这一分布,采用支持向量机、决策树、贝叶斯等多种模型按兴趣主题对用户进行分类.对于基于拓扑关系的分类,依据相同兴趣主题的用户倾向于拥有共同的粉丝这一发现,构建分类模型来按兴趣主题对用户进行分类.然后提出综合信息内容和拓扑关系的分类方法来对用户进行分类.最后基于大规模Twitter数据的实验发现,采用综合方法对用户进行的兴趣分类性能明显高于采用单一信息内容或粉丝拓扑方法的性能.     

Perceived bridging and bonding social capital on Twitter: Differentiating between followers and followees

The present study investigates the influence of Twitter use and the number of followers and followees on perceived bridging and bonding online social capital. Data from a convenience sample of Twitter users (N = 264) indicate that bonding social capital is associated with the number of followers whereas bridging social capital is influenced by the number of followees. Thus, the directed friendship model on Twitter affects different forms of social capital differently. In addition, the study found a negative curvilinear effect of the number of followees on bridging and the number of followers on bonding online social capital. This indicates that the number of followees/followers has positive effects on online bridging/bonding social capital, but only to a certain point. The paper concludes with a discussion of the results in light of theoretical considerations and of implications for future research on the effects of Twitter on social capital.     

The state-of-the-art in personalized recommender systems for social networking

With the explosion of Web 2.0 application such as blogs, social and professional networks, and various other types of social media, the rich online information and various new sources of knowledge flood users and hence pose a great challenge in terms of information overload. It is critical to use intelligent agent software systems to assist users in finding the right information from an abundance of Web data. Recommender systems can help users deal with information overload problem efficiently by suggesting items (e.g., information and products) that match users’ personal interests. The recommender technology has been successfully employed in many applications such as recommending films, music, books, etc. The purpose of this report is to give an overview of existing technologies for building personalized recommender systems in social networking environment, to propose a research direction for addressing user profiling and cold start problems by exploiting user-generated content newly available in Web 2.0.     

大规模社交网络中高效的关键用户选取方法

针对大规模社交网络及其用户发布消息的历史数据,如何快速有效地选取具有较强信息传播能力的关键用户,提出了一种关键用户选取方法。首先,利用社交网络的结构信息,构建以用户为节点的有向图,利用用户发布消息的历史数据,基于Spark计算框架,定量计算由用户活跃度、转发交互度和信息量占比刻画的权重,从而构建社交网络的有向带权图模型;然后,借鉴PageRank算法,建立用户信息传播能力的度量机制,给出基于Spark的大规模社交网络中用户信息传播能力的计算方法;进而,给出基于Spark的d-距选取算法,通过多次迭代,使得所选取的不同关键用户的信息传播范围尽量少地重叠。建立在新浪微博数据上的实验结果表明,所提方法具有高效性、可行性和可扩展性,对于控制不良突发信息传播、社交网络舆情监控具有一定的支撑作用。     

基于文本引导的注意力图像转发预测排序网络

转发预测在社交媒体网站(Social media sites, SMS)中是一个很有挑战性的问题. 本文研究了SMS中的图像转发预测问题, 预测用户再次转发图像推特的图像共享行为. 与现有的研究不同, 本文首先提出异构图像转发建模网络(Image retweet modeling, IRM), 所利用的是用户之前转发图像推特中的相关内容、之后在SMS中的联系和被转发者的偏好三方面的内容. 在此基础上, 提出文本引导的多模态神经网络, 构建新型多方面注意力排序网络学习框架, 从而学习预测任务中的联合图像推特表征和用户偏好表征. 在Twitter的大规模数据集上进行的大量实验表明, 我们的方法较之现有的解决方案而言取得了更好的效果.     

微博中用户标签的研究

微博环境中用户可以为自己添加标签,用户所添加的标签往往被视为是对自身特点和兴趣的重要描述信息.标签中所包含的信息可能有助于建立精确的用户描述,因此在个性化推荐、专家检索、影响力分析等应用中有潜在的应用价值.首先,在大规模数据上分析和研究了微博中用户添加标签的行为及标签内容分布的特点;之后,通过主题模型对用户的微博内容进行分析,实验结果表明:用户的标签越相似,微博内容也越相似,反之亦然;随后,分析了用户关注关系与微博和标签内容之间的联系,实验结果显示,有关注关系的用户之间微博和标签的内容越相似;基于这个发现,分别使用标签内容和微博内容对真实微博数据中的用户关注关系进行预测,结果表明:基于标签的预测方法其效果明显优于基于微博内容的预测方法,显示出用户标签在描述用户兴趣方面的价值.     

微博社交网络的对称程度实证分析

Twitter和Sina微博注册用户构成关注关系社交网络,运用一种对称程度来研究其对称性随社交圈子规模变化的规律。首先根据收集的100万条新浪用户之间的关注关系和236个Twitter用户及其之间的关注关系来构建初始社交网络,选取其中具有明显对称性的连通子网络作为研究的主要对象,通过去除法得到:影响社交网络最大连通子网络对称性的主要因素是大V用户和可忽略用户。其次,采用比较分析法得出Twitter的大V用户构成的社交子网络对称性较强。最后,从功能定位方面分析了两种微博的不同;通过对初始网络的所有连通子网络的对称程度的研究,得出社交圈规模越小、相应的对称性越强的结论。     

面向移动终端的微博信息推荐方法

微型博客(简称“微博')以其简洁方便的交互方式,受到越来越多手机用户的喜爱。然而,微博数据量大、更新速度快以及手机屏幕小、登录网络服务速度较慢等原因,使得用户很难通过移动终端快速了解到近期内微博流行内容。提出一种基于相关主题模型(correlated topic model)的移动微博信息推荐方法,并基于此方法设计了一个可视化移动信息推荐系统。通过‘用户一主题一词语’三维关联矩阵的建立,帮助用户快速了解最近一段时间内的热点主题,并查找与其感兴趣主题相关的其他用户作为备选好友,同时计算主题之间的关联关系,进行主题扩展。在微博代表性网站—Fricndfccd数据集上进行的实验表明了该方法在移动微博信息推荐中的简洁性和有效性。     

Building Ontology for Different Emotional Contexts and Multilingual Environment in Opinion Mining

With the explosive growth of various social media applications, individuals and organizations are increasingly using their contents (e.g. reviews, forum discussions, blogs, micro-blogs, comments, and postings in social network sites) for decision-making. These contents are typical big data. Opinion mining or sentiment analysis focuses on how to extract emotional semantics from these big data to help users to get a better decision. That is not an easy task, because it faces many problems, such as different context may make the meaning of the same word change variously, at the same time multilingual environment restricts the full use of the analysis results. Ontology provides knowledge about specific domains that are understandable by both the computers and developers. Building ontology is mainly a useful first step in providing and formalizing the semantics of information representation. We proposed an ontology DEMLOnto based on six basic emotions to help users to share existed information. The ontology DEMLOnto would help in identifying the opinion features associated with the contextual environment, which may change along with applications. We built the ontology according to ontology engineering. It was developed on the platform Protégé by using OWL2.     

Real-time Internet news browsing: Information vs. experience-related gratifications and behaviors

This study utilized lab observations with 49 subjects to observe what users encounter and how users behave in real-time Internet news browsing. We analyzed users’ selection of news platform, exposure to different topics of news content, and usage of different presentation elements by coding the screen videos. In addition, survey data with the subjects allow us to explore the links between gratifications and Internet news browsing behaviors. Our analyses suggest that users exert their control through actively and selectively interacting with the news services at the platform, content and presentation level to fulfill their different gratifications. In specific, gratifications based on information utility and those based on usage experience show different relations with different kinds of news browsing behaviors. Both the theoretical and methodological contributions are discussed at the end of this paper.     

All the news that’s fit to post: A profile of news use on social networking sites

Facebook and other social networking sites (SNSs)¹ are altering the way individuals communicate. These online environments allow users to keep up with friends, network with colleagues, and share their personal views and observations with others. Previous work describes typical social networking site users as young, extroverted, and technologically savvy. Little research exists, however, on the emerging role of news in the social network environment. With over 500 million global Facebook users, both print and television based media outlets are making concerted efforts to become part of this important and increasingly ubiquitous virtual world. The present study uses a sample of students, faculty, and staff from a large university to investigate the factors that are related to news use on Facebook. Findings indicate that while news use is still a minor component of overall social network site activity, certain key variables, such as gender and life satisfaction, have a significant impact on how Facebook is used for news-related purposes. Future implications for news in the social networking world are presented and discussed.     

面向微博客的热点事件情感分析方法

微博客是一种新兴的网络信息交互平台,近年来受到越来越多的用户的关注。信息的简洁性以及传播渠道的多样性使得微博客成为广大网民浏览热点事件相关信息和发表个人观点的重要途径。分析和监测微博客内容中所包含的情感信息,能够了解民众对特定热点事件的关注程度和情感变化,从而辅助评佑和掌握事件的发展状况。因此,提出一种面向微博客的热点事件情感分析方法,该方法首先自动挖掘用户对某热点事件的多个关注点,并针对不同关注点进行情感分析以及情感趋势监测,最终实现一个可视化的热点事件情感趋势分析原型系统。通过实例验证了微博客信息在网络热点事件的情感分析和监测中的有效性。     

How Can Our Tweets Go Viral? Point-Process Modelling of Brand Content

People create and share content via online social networks, which provide an unparalleled opportunity for brands to gain visibility, promote products or services and drive revenue growth. Much research has focused on why, how, or what social content is popular, trending and “hype”. One central challenge is to forecast the spread (cascades) of information that leads to the popularity of content throughout a social network. Online content tends to have bursts and spikes, experiencing a different cascading pattern depending on the viral propagation. In this paper, we propose and test a flexible framework capable of modelling such patterns and trends. We take temporal and network perspectives and develop a model based on the multivariate Hawkes processes that account for social behaviour and network elements such as follower counts, and activity variation observed in collective re-sharing behaviour. We focus on Twitter as the most widely used micro-blogging online social network and measure the popularity of a brand's tweet by analysing the time-series path of three types of subsequent activities (retweets (RTs), replies (REs) and likes (LKs)). The specific model that we propose in this paper is the multidimensional epidemic-type aftershock sequence (METAS) model, a particular case of the multivariate Hawkes process. It consists of a power-law relaxation governing the timing of activities. It also includes an exponential boost as a reinforcement mechanism for the response amplitude to model the impact of influential users on their followers. Earlier attempts to model online cascades have treated all online responses as one type of activity. Rather than aggregating all the activities into one stream, and therefore, ignoring exciting effects among different types of activities, we incorporate the activity variation into the predictive models of content popularity, explicitly accounting for such excitation effects. We develop epidemic-type mutually exciting Hawkes point processes models to quantify such effects and to predict more accurately the number of follow-up activities (i.e., RTs, REs and LKs) on a brand tweet after it is posted. Our results suggest that the proposed model outperforms the state-of-the-art models in terms of prediction accuracy, as it is able to account for mutual excitations and cross-interactions between sequences of users’ activities from one type to another. These results are relevant for developing and executing a plan for online activities by the brand owners.     

Social network technologies for semantic linking of information objects in scientific digital library

In the last decade, scientific digital libraries were traditionally used for publishing research results and for enabling wide open access to them. Functional capabilities of digital libraries can be extended by offering users the opportunity of linking information objects of the library and providing for created linkages explicitly defined semantics based on a given ontology. Such an activity of users, which is peculiar to social networks, motivated by different reasons, and carried out on their own initiative, results in the dynamic semantic structure of the digital library content. In the environment of such a kind of a social network, certain new forms of scientific activities become possible and data sources can be created that provide more information for scientometric researches as compared to presently available ones. In this paper, we propose an approach for creating such networks and discuss results of its implementation in the Socionet environment; the Socionet is a large-scale online information space that covers information resources of a number of scientific, educational, etc. organizations.