Thesis (Ph. D.)--University of Texas at Austin, 2008. / Vita. Includes bibliographical references.

Thesis (Ed. D.)--UCLA, 2008. / Description based on print version record. Includes bibliographical references (leaves 124-132).

M.Sc. (Computer Science) / With the phenomenal growth of the Online Social Network (OSN) industry in the past few years, users have resorted to storing vast amounts of personal information on these sites. The information stored on these sites is often readily accessible from anywhere in the world and not always protected by adequate security settings. As a result, user information can make its way, unintentionally, into the hands of not only other online users, but also online abusers. Online abusers, better known as cyber criminals, exploit user information to commit acts of identity theft, Advanced Persistent Threats (APTs) and password recovery, to mention only a few. As OSN users are incapable of visualising the process of access to their OSN information, they may choose to never adjust their security settings. This can become synonymous with ultimately setting themselves up to becoming a victim of cyber crime. In this dissertation we aim to address this problem by proposing a prototype system, the Information Deduction Model (IDM) that can visualise and simulate the process of accessing information on an OSN profile. By visually explaining concepts such as information access, deduction and leakage, we aim to provide users with a tool that will enable them to make more informed choices about the security settings on their OSN profiles thereby setting themselves up for a pleasant online experience.

Traditional marketing models are swiftly being upended by the advent of online social networks. Yet, practicing firms that are engaging with online social networks neither have a reliable theory nor sufficient practical experience to make sense of the phenomenon. Extant theory in particular is based on observations of the real world, and may thus not apply to online social networks. Practicing firms may consequently be misallocating a large amount of resources, simply because they do not know how the online social networks with which they interact are organized. The purpose of this dissertation is to investigate how online social networks that are in stark contrast to real-world social networks behave and how they get organized. In particular, I explore how network structure and information flow within the network impact each other, and how they affect the phenomenon of influence in online social networks. I have collected retrospective data from Twitter conversations about six YouTube product categories (Music, Entertainment, Comedy, Science, Howto and Sports) in continuous time for a period of three months. Measures of network structure (Scale Free Metric, Assortativity and Small World Metric), network flows (Total Paths, Total Shortest Paths, Graph Diameter, Average Path Length, and Average Geodesic Length) and influence (Eigenvector Centrality/Centralization) were computed from the data. Experimental measures such as power law distributions of paths, shortest paths and nodal eigenvector centrality were introduced to account for node-level structure. Factor analysis and regression analysis were used to analyze the data and generate results. The research conducted in this dissertation has yielded three significant findings. 1. Network structure impacts network information flow, and conversely; network flow and network structure impact the network phenomenon of influence. However, the impact of network structure and network flow on influence could not be identified in all instances, suggesting that it cannot be taken for granted. 2. The nature of influence within a social network cannot be understood just by analyzing undirected or directed networks. The behavioral traits of individuals within the network can be deduced by analyzing how information is propagated throughout the network and how it is consumed. 3. An increase or decrease in the scale of a network leads to the observation of different organizational processes, which are most likely driven by very different social phenomena. Social theories that were developed from observing real-world networks of a relatively small scale (hundreds or thousands of people) consequently do not necessarily apply to online social networks, which can exhibit significantly larger scale (tens of thousands or millions of people). The primary contribution of this dissertation is an enhanced understanding of how online social networks, which exhibit contrasting characteristics to social networks that have been observed in the real world, behave and how they get organized. The empirical findings of this dissertation may allow practicing managers that engage with online social networks to allocate resources more effectively, especially in marketing. The primary limitations of this research are the inability to identify the causes of change within networks, glean demographic information and generalize across contexts. These limitations can all be overcome by follow-on studies of networks that operate in different contexts. In particular, further study of a variety of online social networks that operate on different social networking platforms would determine the extent to which the findings of this dissertation are generalizable to other online social networks. Conclusions drawn from an aggregation of these studies could serve as the foundation of a more broadly-based theory of online social networks.

published_or_final_version / Journalism and Media Studies Centre / Master / Master of Philosophy

A dissertation submitted to the Faculty of Commerce, Law and Management, University of the Witwatersrand, Wits Business School to fulfilments of the requirements of a Master of Management by Research 30 March 2015 / Our understanding of the psychological construct of sense of community has been developing steadily, particularly since the publication of the seminal work by McMillan and Chavis (1986). Pertinent to this study, the sense of community construct has been applied to the virtual world, leading to the concept of sense of virtual community (SoVC), that is, a sense of community felt by members of a virtual community. This study synthesises the findings of the extant literature to build a multi-dimensional model of sense of community. Moving to a specific context, this study examines SoVC among members of cause-related virtual communities. Examples of such communities can be found in the Facebook communities that have developed around the various branches of the Red Cross and of World Vision. Among members of such communities, some level of support for the mediating cause organisation can be presumed to exist. This is referred to in this dissertation as Passion for the Cause (PFC). Empirical and theoretical work on the interaction between SoVC and PFC is lacking. This study investigates the extent to which SoVC and PFC are associated and seeks to bring clarity to the nature of the association. The research instrument was an online self-completion survey. The Facebook pages of South African cause organisations were used to invite community members to complete the survey. Respondents were participants in the Facebook-based communities of South African cause organisations (n = 67). The research instrument included a scale for SoVC (12 items) and a scale for PFC (6 items). An exploratory factor analysis was done to identify the latent factors of SoVC in this context. Adequate support was found for the conceptualisation of three factors of SoVC, namely, General Benefit, Friendship, and Helping. This was followed by a series of multiple regression analyses aimed at testing the relationships between PFC and SoVC and its factors. SoVC and PFC were found to be highly correlated. Furthermore, PFC was found to significantly predict SoVC. It was also found to predict the SoVC factor conceptualised as General Benefit. Finally, SoVC was found to predict PFC. Notably, PFC was found to be less able to predict SoVC than was SoVC able to predict PFC. Implications for the moderators of cause-related virtual communities are discussed. / MB2016

Orientador: Sandro Caramaschi / Banca: Marco Antonio Corrêa Varella / Banca: Erico Bruno Viana Campos / Resumo: Os aplicativos de relacionamento se tornaram populares e pertencentes ao cotidiano dos indivíduos. Dentre eles, existem aqueles destinados exclusivamente aos desdobramentos amorosos. Apesar de serem utilizados amplamente, pouco se sabe cientificamente sobre seu uso e com isso, pretendeu-se a partir desse trabalho levantar dados a respeito de sua utilização para a seleção de parceiros. Diante disso, o objetivo dessa pesquisa foi verificar a dinâmica amorosa que ocorre por meio de tais aplicativos, as características sociais atribuídas aos seus usuários, quais são as vantagens e desvantagens apontadas pelos participantes em relação ao seu uso, se os usuários estão satisfeito com os desdobramentos, se possuem baixo ou alto conservadorismo sexual e autoestima e quais são as características que procuram em um parceiro considerado ideal para ficar e para casar. Para isso, realizou-se a comparação entre os participantes de ambos os sexos, a fim de verificar se existem diferenças estatisticamente significativas entre homens e mulheres. Para tanto, 210 universitários matriculados em diferentes cursos de uma mesma instituição de ensino participaram da pesquisa respondendo presencialmente a um questionário. Constataram-se diferenças significativas entre os sexos no que diz respeito ao número de pessoas por quem se sentem atraídos em seu círculo social e o número de encontros que concretizaram com os pretendentes virtuais, de modo que a amostra masculina superou numericamente a feminina.... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Relationship applications have become popular and pertaining to individuals' daily lives. Among them, there are those destined exclusively for the unfolding of love. Although widely used, little is known scientifically about its use and with that, it was intended from this work to collect data regarding its use for the selection of partners. Therefore, the objective of this research was to verify the amorous dynamic that occurs through such applications, the social characteristics attributed to its users, what are the advantages and disadvantages pointed out by the participants in relation to their use, if the users are satisfied with the unfolding, whether they have low or high sexual conservatism and self-esteem, and what are the characteristics they are looking for in a partner considered ideal for staying and getting married. For this, the comparison between the participants of both sexes was carried out in order to verify if there are statistically significant differences between men and women. For that, 210 university students enrolled in different courses of the same teaching institution participated in the research, responding face to face to a questionnaire. Significant differences were found between the sexes in terms of the number of people they are attracted to in their social circle and the number of encounters they made with virtual pretenders, so that the male sample surpassed the female numerically. The characteristics that the participants attributed to users... (Complete abstract click electronic access below) / Mestre

社會意見在塑造我們購買決策和購買經歷發揮了至關重要的作用。除了正面（或負面）的意見會鼓勵（或打消）我們購買某個產品，我們的意見更傾向於遵循我們的社交圈。社會意見的這些方面對於做出精確產品推薦、準確預測信息流向、及有效營銷活動發布極為重要。 / 在這篇論文中，我們首先研究極性意見對我們的購買決策的影響。同時，我們分析了兩個現實世界中的社會網絡，Flixster 和Epinions 中的消極和積極的意見的信息傳播模式。我們觀察到，否定意見的存在大大降低了表達意見的數量。考慮到這兩種意見的不對稱性，我們提出並擴展了目前最流行的兩個信息傳播模式，獨立分級和線性閾值模型。我們提出的拓展模型提供了一個可處理的影響問題和並能夠提高將來意見的預測精度，超過3%。 / 更進一步，我們研究了社會意見對我們表達產品意見的影響。該問題的假設是多次顯示我們表達的意見並不完全獨立於我們的社交圈，而是通過校準，使之跟社會意見相似。為了理解這一現象，我們為用戶的評分提出了一個新型的模型。該模型中，用戶對項目的評分是由社會輿論、用戶的偏好和項目特點的一個函數。該模型可以提高用戶評分的預測準確率達2%。此外，模型中學習到的參數可展示用戶對社會意見的遵循程度。用戶對社會意見的遵循分析表明，超過76%的用戶傾向於在一定程度上遵循他們好友的意見。平均而言，當社會影響存在的時候，用戶評分更趨於正面。我們還發現，社會的遵循者通常不是信息傳播的第一次參與者。 / Social opinions play a crucial role in shaping both our purchase decisions and our experience. While on one hand, we are encouraged (discouraged) to adopt a product upon hearing the positive (negative) opinions; on the other hand, our opinions tend to conform to our social circle. Both of these aspects of social opinions are important in order to make precise product recommendations, to accurately predict the information flow pathways and to launch efficient viral marketing campaigns. / In this thesis, we first study the impact of polarity of opinions on our purchase decisions. For the same, we analyze the information propagation patterns of the negative and positive opinions on two real world social networks, Flixster and Epinions, and observe that the presence of negative opinions significantly reduces the number of expressed opinions. To account for the asymmetry between the two kinds of opinions, we propose extensions of the two most popular information propagation models, Independent Cascade and Linear Threshold models. The proposed extensions give a tractable influence problem and improve the prediction accuracy of future opinions, by more than 3%. / Next, we study the impact of social opinions on our expressed opinions about the products. The hypothesis is that many times our expressed opinions are not completely independent of our social circle and gets calibrated such that they are similar to the social opinions. In order to understand this phenomenon, we propose a novel formulation for the users ratings where every expressed rating is considered as a function of the social opinion along with the user preference and item characteristics. The proposed method helps in improving the prediction accuracy of users’ rating by more than 2% in presence of social influence. Additionally, the learned model parameters reveal the degree of conformity of users. Detailed analysis of user social conformity show that more than 76% of users tend to conform to their friends to some extent. On an average, user ratings become more positive in presence of the social influence. We also nd that the social conformers are usually not the rst one to participate in an information cascade. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Garg, Priyanka. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 100-110). / Abstracts also in Chinese. / Abstract --- p.i / Acknowledgement --- p.v / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Contributions --- p.3 / Chapter 1.2 --- Organization --- p.4 / Chapter 2 --- Background & Survey --- p.6 / Chapter 2.1 --- Network Structure --- p.6 / Chapter 2.1.1 --- Basic Definitions --- p.6 / Chapter 2.1.2 --- Structural Properties of Social Networks --- p.8 / Chapter 2.1.3 --- Network Generators --- p.12 / Chapter 2.2 --- Information Diffusion in Social Networks --- p.16 / Chapter 2.2.1 --- Basic Terminologies --- p.19 / Chapter 2.2.2 --- Principles governing the Decision-Making process --- p.19 / Chapter 2.2.3 --- Information Cascade Models --- p.21 / Chapter 2.2.4 --- Influence Estimation --- p.34 / Chapter 2.2.5 --- Viral Marketing --- p.39 / Chapter 2.2.6 --- Influence vs. Homophily --- p.44 / Chapter 2.2.7 --- Results from Large Scale Empirical Studies --- p.45 / Chapter 3 --- Impact on Product Purchase Decision --- p.47 / Chapter 3.1 --- Introduction --- p.47 / Chapter 3.2 --- Related Work --- p.49 / Chapter 3.3 --- Problem Definition --- p.50 / Chapter 3.4 --- Data and Observations --- p.51 / Chapter 3.4.1 --- Data Collection --- p.51 / Chapter 3.4.2 --- Observations --- p.52 / Chapter 3.5 --- Polarity-Sensitive Information Flow Model --- p.54 / Chapter 3.5.1 --- Social Influence Function --- p.55 / Chapter 3.5.1.1 --- Polarity-Sensitive IC Model --- p.55 / Chapter 3.5.1.2 --- Polarity-Sensitive LT Model (LTPS) --- p.58 / Chapter 3.5.2 --- Activation State of Influenced Node --- p.59 / Chapter 3.6 --- Influence Estimation --- p.61 / Chapter 3.7 --- Experiments on Synthetic Data --- p.63 / Chapter 3.7.1 --- IA and WP as Approximation of IC-N --- p.64 / Chapter 3.7.2 --- Quality of the Estimated Parameters --- p.65 / Chapter 3.7.3 --- Prediction Accuracy --- p.66 / Chapter 3.8 --- Experiments on Real Data --- p.69 / Chapter 3.8.1 --- Experimental Setup --- p.69 / Chapter 3.8.2 --- Observations --- p.70 / Chapter 3.9 --- Summary --- p.72 / Chapter 4 --- Impact on Posterior Evaluation --- p.73 / Chapter 4.1 --- Introduction --- p.73 / Chapter 4.2 --- Related Work --- p.75 / Chapter 4.3 --- Ratings under social conformity --- p.77 / Chapter 4.3.1 --- Problem Definition & Notations --- p.79 / Chapter 4.3.2 --- Conformer's Ratings --- p.80 / Chapter 4.3.3 --- Parameter Estimation --- p.82 / Chapter 4.4 --- Evaluation --- p.84 / Chapter 4.4.1 --- Goodreads Dataset --- p.85 / Chapter 4.4.2 --- Prediction Accuracy --- p.87 / Chapter 4.4.3 --- Influencers Quality --- p.89 / Chapter 4.5 --- Social Conformity Analysis --- p.91 / Chapter 4.6 --- Summary --- p.96 / Chapter 5 --- Summary & Future Work --- p.97 / Chapter 5.1 --- Summary --- p.97 / Chapter 5.2 --- Future Work --- p.98 / Bibliography --- p.100

当今，越来越多的用倾向于使用论坛、博客、脸书等在线工具来表达关于商品、电影和政治等话题的观点。这些观点不仅可以帮助用进行决策，同时也为各个商业和社会领域提供了具有重要价值的反馈信息。因此，面向观点应用成为了当前最活跃的研究领域之一，其中包括观点检索，观点摘要，观点问答。面向观点应用与面向事实应用的根本区别是信息需求的不同，分别是传统的客观信息和主观信息。所谓主观信息是指对于某个特定目标的观点或评论。为了表示主观信息，应该综合考虑观点性、主题相关性，以及观点与主题之间的关联。现有的基于词袋的表示方法将词作为描述客观信息的基本语义单元，它可以有效的表示主题相关性以满足客观信息的需求。而主观信息需要同时考虑观点性和主题相关性，由于单独一个词不能同时表示观点性和相关性，因此词不再是最小的语义单位。此外，基于词袋的表示方法忽略了词序和词义，这使得观点性和相关性两类信息通常混在一起，难以区分。因此，基于词袋方法不能够准确的表示主观信息，并严重的影响了面向观点应用的性能。 / 本文回答了以下几个由主观信息表示不当所引发的研究问题： 1. 对于主观信息而言单个词将不再是基本语义单元，是否存在一种有效的表示方法对其进行描述？ 2. 由于主观信息是观点信息和相关性信息的结合，如何利用新的表示方法来描述这二者之间的关联信息？3. 如何对主观信息进行量化，以便对文档进行检索和分析？ 4. 如何在面向观点应用中实现全新的主观信息表示方法？ / 由于观点检索的结果会直接影响到其它面向观点应用的性能，因此本文从观点检索这一问题入手。首先，我们提出了一种基于句子的方法来分析词袋表示方法的局限性。以此为据，定义了一种具有丰富语义的表达方式来表示主观信息，即词对，它是由出现在同一句子中的情感词和与之关联的目标词共同组成的。然后，我们提出了一系列方法来描述和获取两类语境信息：1）观点内信息：我们给出了三种提取词对的方法以获取观点与主题的关联信息；2）观点间信息：我们提出了一种权重计算方法来度量词对间的相关程度，从而获取词对与词对之间的关系。最后，我们集成了观点内信息和观点间信息并提出了潜在情感关联模型来解决观点检索这一问题。在标准数据集上的实验结果表明，基于词对的表示方法可以有效地描述主观信息，同时潜在情感关联模型能够获取词与词之间的关联信息，从而实现了利用语境信息提高观点检索的效果。 / 此外，我们将词对应用于观点摘要和观点问答中，标准数据集上的评测结果显示基于词对的主观信息表示方法对于其它面向观点应用也同样有效。 / There is a growing interest for users to express their opinions about products, films, politics, by using on-line tools such as forums, blogs, facebooks, etc. These opinions cannot only help users make decisions, e.g., whether to buy a product, but also to ob-tain valuable feedback for business and social events. Today, research on opin-ion-oriented applications (OOAs) including opinion retrieval, opinion summarization and opinion question and answering is attracting much attention. The difference be-tween fact-based and opinion-oriented applications lies in users‘ information need. The former requires objective information and the latter subjective, which comprises of opinions or comments expressed on a specific target. To meet the need of subjective information, both opinionatedness and relevance together with the association between them should be taken into account. Existing systems represent documents in bag-of-word. However, this representation fails to distinguish opinionatedness from relevance. Moreover, due to the ignorance of word sequence, words associations are lost. For this reason, bag-of-word representation is ineffective for subjective information, and affects the performance of OOAs seriously. / In this thesis, we try to answer the following challenging questions arose in subjective information representation. Since word is no longer the basic semantic unit, how would subjective information be represented? Subjective information is a combination of opinionatedness and relevance, so how would the association between them be modeled? How would subjective information be measured for the purpose of document ranking, retrieval, and analysis? How would opinion-oriented applications benefit from subjective information? / We start from solving the problem of opinion retrieval whose results can directly influence the performance of other opinion-oriented applications. We first present a sentence-based approach to analyze the limitation of bag-of-word representation and define a semantically richer representation, namely word pair for subjective infor-mation. A word pair is constructed by a sentiment word and its associated target co-occurring in a sentence. We then propose techniques to capture two kinds of con-textual information. 1) Intra-opinion information: three methods are proposed to ex-tract the word pair. 2) Inter-opinion information: a weighting scheme is present to measure the weight of individual word pair. Finally, we devise an algorithm to integrate both intra-opinion and inter-opinion information into a latent sentimental association model for opinion retrieval. The evaluation on three benchmark datasets suggests the effectiveness of word pair and the latent sentimental association retrieval model provide insight into the words association to support opinion retrieval beneficial from pairwise representation. We also apply word pair to opinion summarization and opinion question answering. The evaluation on two benchmark datasets shows that word pair performs effectively in the applications. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Li, Binyang. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references (leaves [96]-103). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese. / Abstract --- p.ii / Abstract in Chinese --- p.iv / Acknowledgements --- p.vi / Contents --- p.viii / List of Tables --- p.xi / List of Figures --- p.xiii / Chapter 1. --- Introduction --- p.1 / Chapter 1.1. --- Problem and Challenges --- p.3 / Chapter 1.1.1 --- Subjective Information Representation --- p.3 / Chapter 1.1.2 --- Associative Information in an Opinion Expression --- p.4 / Chapter 1.1.3 --- Opinion Expression Measurement --- p.5 / Chapter 1.1.4 --- Applications of Subjective Information Representation to Different OOAs --- p.6 / Chapter 1.2. --- Contributions --- p.6 / Chapter 1.3. --- Chapter Summary --- p.7 / Chapter 2. --- Pairwise Representation --- p.9 / Chapter 2.1 --- Related Woks on Opinion Retrieval --- p.10 / Chapter 2.1.1 --- Opinion Retrieval Models --- p.10 / Chapter 2.1.2 --- Lexicon-based Opinion Identification --- p.12 / Chapter 2.2 --- Sentence-based Approach for Opinion Retrieval --- p.13 / Chapter 2.2.1 --- The Limitations of Document-based Approaches for Opinion Retrieval --- p.13 / Chapter 2.2.2 --- Sentence-based Approach for Opinion Retrieval --- p.16 / Chapter 2.2.3 --- Evaluation and Results --- p.21 / Chapter 2.2.4 --- Summary --- p.26 / Chapter 2.3 --- Pairwise Representation --- p.28 / Chapter 2.3.1 --- Definition of Word Pair --- p.28 / Chapter 2.3.2 --- Sentiment Lexicon Construction --- p.29 / Chapter 2.3.3 --- Topic Term Lexicon Construction --- p.30 / Chapter 2.3.4 --- Word Pair Construction --- p.31 / Chapter 2.4 --- Graph-based Model for Opinion Retrieval --- p.33 / Chapter 2.4.1 --- HITS Model for Opinion Retrieval --- p.34 / Chapter 2.4.2 --- PageRank Model for Opinion Retrieval --- p.37 / Chapter 2.4.3 --- Evaluation and Results --- p.40 / Chapter 2.5 --- Chapter Summary --- p.50 / Chapter 3. --- Pairwise Representation Measurement --- p.51 / Chapter 3.1 --- Word Pair Weighting Scheme --- p.52 / Chapter 3.1.1 --- PMI-based Weighting Scheme --- p.52 / Chapter 3.1.2 --- Evaluation and Results --- p.56 / Chapter 3.1.3 --- Summary --- p.60 / Chapter 3.2 --- Latent Sentimental Association --- p.61 / Chapter 3.2.1 --- Problem Formulation --- p.61 / Chapter 3.2.2 --- LSA Integrated Generative Model --- p.62 / Chapter 3.2.3 --- Modeling the Dependency between Q and d --- p.64 / Chapter 3.2.4 --- Modeling the Dependency between O and d --- p.67 / Chapter 3.3 --- Parameter Estimation --- p.67 / Chapter 3.3.1 --- Estimating P(Q --- p.67 / Chapter 3.3.2 --- Estimating MI(Q,O --- p.69 / Chapter 3.4 --- Evaluation and Results --- p.69 / Chapter 3.5 --- Chapter Summary --- p.72 / Chapter 4. --- Pairwise Representation in Opinion-oriented Application --- p.75 / Chapter 4.1. --- Opinion Questioning and Answering --- p.76 / Chapter 4.1.1 --- Problem Statement --- p.76 / Chapter 4.1.2 --- Existing Solution --- p.78 / Chapter 4.1.3 --- A Word Pair based Approach for Sentence Ranking --- p.79 / Chapter 4.1.4 --- Answer Generation --- p.82 / Chapter 4.1.5 --- Evaluation and Results --- p.82 / Chapter 4.2. --- Opinion Summarization --- p.86 / Chapter 4.2.1 --- Problem Statement --- p.86 / Chapter 4.2.2 --- Existing Solution --- p.87 / Chapter 4.2.3 --- Sentence Ranking --- p.88 / Chapter 4.2.4 --- Summary Generation --- p.88 / Chapter 4.2.5 --- Evaluation and Results --- p.89 / Chapter 4.3. --- Chapter Summary --- p.91 / Chapter 5. --- Conclusions and Future Works --- p.93 / Bibliography --- p.97

Online health communities (OHCs) have been increasingly popular among patients with chronic or life-threatening illnesses for the exchange of social support. Contemporary research of OHCs relies on methods and tools to handle analytics of massive user-generated content at scale to complement traditional qualitative analysis. In this thesis, we aim at advancing the area of research by providing computational tools and methods which facilitate automated content analysis, and by presenting applications of these tools to investigating member characteristics and behaviors. We first provide a framework of conceptualization to systematically describe problems, challenges, and existing solutions for OHCs from a social support standpoint, to bridge the knowledge gap between health psychology and informatics. With this framework in hand, we define the landscape of online social support, summarize current research progress of OHCs, and identify research questions to investigate for this thesis. We then build a series of computational tools for analyzing OHC content, relying on techniques of machine learning and natural language processing. Leveraging domain-specific features, our tools are tailored to handle content analysis tasks on OHC text effectively. Equipped with computational tools, we demonstrate how characteristics of OHC members can be identified at scale in an automated fashion. In particular, we build up multi-dimensional descriptions for patient members, consisting of what topics they focus on, what sentiment they express, and what treatments they discuss and adopt. Patterns of how these member characteristics change through time are also investigated longitudinally. Finally, relying on computational analytics, members' behaviors of engagement such as debate and dropping-out are identified and characterized. Studies presented in this thesis discover static and longitudinal patterns of member characteristics and engagement, which are potential research hypotheses to be explored by health psychologists and clinical researchers. The thesis also contributes to the informatics community by making computational tools, lexicons, and annotated corpora available to facilitate future research.