Tato diplomová práce navazuje na bakalářskou práci, ve které byl vytvořen model důvěry pro sociální síť Facebook. Do tohoto modelu jsou zapracovány připomínky z konference UMAP 2013 a ověřena jeho škálovatelnost a flexibilita. V další části práce jsou uvedeny základní termíny z psychologie osobnosti a zkoumána závislost důvěry na osobnosti uživatele. Je vybrán model Big Five k reprezentaci charakteru uživatele a navržen dotazník, u nějž bude zkoumána korelace s modelem důvěry. Tato korelace je na základě sociologických poznatků odhadnuta a později ověřena na reálných uživatelích sociální sítě Facebook.

This diploma thesis is focusing on description of processing social network analysis, design and implementation of a model that simulates a particular social network and its analysis. Social networks are modern and very used in this time. They are very good point for exploring. This project deal with static analysis social network, where social network is constructed by graph. We nd out di erent properties of single component and than we establish signi cance of them. Relationships between components are important too for us, because they have a big influence on propagation information in network. Structural properties figure out existence of di fferent communities. We simulate social network with multi-agent systems, they are desirable for represent changes in network. Multi-agent systems have implemented a simulation model that represents a particular social network. His behaviour was analyzed and examinated by chosen methods.

<div>Representation learning (RL) for social networks facilitates real-world tasks such as visualization, link prediction and friend recommendation. Many methods have been proposed in this area to learn continuous low-dimensional embedding of nodes, edges or relations in social and information networks. However, most previous network RL methods neglect social signals, such as textual communication between users (nodes). Unlike more typical binary features on edges, such as post likes and retweet actions, social signals are more varied and contain ambiguous information. This makes it more challenging to incorporate them into RL methods, but the ability to quantify social signals should allow RL methods to better capture the implicit relationships among real people in social networks. Second, most previous work in network RL has focused on learning from homogeneous networks (i.e., single type of node, edge, role, and direction) and thus, most existing RL methods cannot capture the heterogeneous nature of relationships in social networks. Based on these identified gaps, this thesis aims to study the feasibility of incorporating heterogeneous information, e.g., texts, attributes, multiple relations and edge types (directions), to learn more accurate, fine-grained network representations. </div><div> </div><div>In this dissertation, we discuss a preliminary study and outline three major works that aim to incorporate textual interactions to improve relational representation learning. The preliminary study learns a joint representation that captures the textual similarity in content between interacting nodes. The promising results motivate us to pursue broader research on using social signals for representation learning. The first major component aims to learn explicit node and relation embeddings in social networks. Traditional knowledge graph (KG) completion models learn latent representations of entities and relations by interpreting them as translations operating on the embedding of the entities. However, existing approaches do not consider textual communications between users, which contain valuable information to provide meaning and context for social relationships. We propose a novel approach that incorporates textual interactions between each pair of users to improve representation learning of both users and relationships. The second major component focuses on analyzing how users interact with each other via natural language content. Although the data is interconnected and dependent, previous research has primarily focused on modeling the social network behavior separately from the textual content. In this work, we model the data in a holistic way, taking into account the connections between the social behavior of users and the content generated when they interact, by learning a joint embedding over user characteristics and user language. In the third major component, we consider the task of learning edge representations in social networks. Edge representations are especially beneficial as we need to describe or explain the relationships, activities, and interactions among users. However, previous work in this area lack well-defined edge representations and ignore the relational signals over multiple views of social networks, which typically contain multi-view contexts (due to multiple edge types) that need to be considered when learning the representation. We propose a new methodology that captures asymmetry in multiple views by learning well-defined edge representations and incorporates textual communications to identify multiple sources of social signals that moderate the impact of different views between users.</div>

Purpose – This thesis aims to investigate the use and design of informal controls inan organization’s management control package (MCP). Informal controls areoperationalized as the intra-organizational social network.  Design/methodology/approach – An inductive case study is used to investigate ahybrid organization. The thesis uses data from interviews, documents, and a surveyto gather information about the organization’s MCP and intra-organizational socialnetwork.  Findings – It was found that the use of informal controls in the case organizationwas related to the spreading of clans and beliefs. The intra-organizational socialnetwork acts as conduits for clans, while the social network at the individual levelrelates to the spreading of certain beliefs. Furthermore, it was found that informalcontrols play a part in the design of different groups’ MCPs. Specifically, thatcertain groups use informal controls differently depending on whether theycoordinate other groups or directly work with the ongoing operations.  Research limitation/implications – This study has implications for the MCPresearch, as it is the first empirical study to operationalize informal controls associal networks in an MCP. It shed light on the role of informal controls in theMCP. A limitation is that no observational data has been used to analyze theorganizational culture, which may have affected the conclusions. Originality/value – This is the first empirical study that operationalizes informalcontrols as social networks in an MCP. The study also presents the MCPs of 38intra-organizational groups and classifies them according to configurations. / Syfte – Syftet med denna uppsats är att undersöka användandet och designen avinformell kontroll, i form av ett intra-organisatoriskt social nätverk, i enorganisations styrpaket. Metod – En induktiv fallstudie har använts för att studera en hybrid-organisation.Studien använder data från intervjuer, dokument och en enkät för att kunnakonstruera organisationens styrpaket och intra-organisatoriska sociala nätverk. Slutsats – Studien visar på att användandet av informell kontroll inomorganisationen är relaterat till sprdiningen av klaner och övertygelser. Det intraorganisatoriska sociala nätverket agerar som ledningar för spridningen av klaner,medan det sociala nätverket mellan individer är relaterat till vissa övertygelser.Utöver detta, så visar även studien på att informell kontroll har en roll i designen avolika gruppers styrpaket. Mer specifikt, att grupper använder informell kontroll påolika sätt beroende på om de kordinerar andra grupper eller själva utför aktiviteter iden dagliga verksamheten.  Begränsingar &amp; Implikationer – Denna uppsats har implikationer för litteraturenom styrpaket. Det är den första empiriska studien som operationaliserar informellkontroll som ett socialt nätverk vid studie av ett styrpaket. Uppsatsen belyser däravrollen av informell kontroll i styrpaketet. En av studiens begränsningar är att ingenobservationsdata har använts för att analysera verksamhetens organisationskultur,detta kan möjligen fått konsekvenser för slutsatsen Originalitet – Detta är den första empiriska studien som operationaliserar informellkontroll som ett social nätverk vid stidue av ett styrpaket. Denna uppsats presenteraräven styrpaketen hos 38 styckna intra-organisatoriska grupper och klassificerar demi enlighet med styrkonfigurationer

No description available.

Master of Science / Department of Computing and Information Sciences / Doina Caragea / The idea of socializing with other people of different backgrounds and cultures excites the web surfers. Today, there are hundreds of Social Networking sites on the web with millions of users connected with relationships such as "friend", "follow", "fan", forming a huge graph structure. The amount of data associated with the users in these Social Networking sites has resulted in opportunities for interesting data mining problems including friendship link and interest predictions, tag recommendations among others. In this work, we consider the friendship link prediction problem and study a topic modeling approach to this problem. Topic models are among the most effective approaches to latent topic analysis and mining of text data. In particular, Probabilistic Topic models are based upon the idea that documents can be seen as mixtures of topics and topics can be seen as mixtures of words. Latent Dirichlet Allocation (LDA) is one such probabilistic model which is generative in nature and is used for collections of discrete data such as text corpora. For our link prediction problem, users in the dataset are treated as "documents" and their interests as the document contents. The topic probabilities obtained by modeling users and interests using LDA provide an explicit representation for each user. User pairs are treated as examples and are represented using a feature vector constructed from the topic probabilities obtained with LDA. This vector will only capture information contained in the interests expressed by the users. Another important source of information that is relevant to the link prediction task is given by the graph structure of the social network. Our assumption is that a user "A" might be a friend of user "B" if a) users "A" and "B" have common or similar interests b) users "A" and "B" have some common friends. While capturing similarity between interests is taken care by the topic modeling technique, we use the graph structure to find common friends. In the past, the graph structure underlying the network has proven to be a trustworthy source of information for predicting friendship links. We present a comparison of predictions from feature sets constructed using topic probabilities and the link graph separately, with a feature set constructed using both topic probabilities and link graph.

Master of Science / Department of Computing and Information Sciences / Doina Caragea / William H. Hsu / An ontology can be seen as an explicit description of the concepts and relationships that exist in a domain. In this thesis, we address the problem of building an interests' ontology and using the same to construct features for predicting both potential friendship relations between users in the social network Live Journal, and users' interests. Previous work has shown that the accuracy of predicting friendship links in this network is very low if simply interests common to two users are used as features and no network graph features are considered. Thus, our goal is to organize users' interests into an ontology (specifically, a concept hierarchy) and to use the semantics captured by this ontology to improve the performance of learning algorithms at the task of predicting if two users can be friends. To achieve this goal, we have designed and implemented a hybrid clustering algorithm, which combines hierarchical agglomerative and divisive clustering paradigms, and automatically builds the interests' ontology. We have explored the use of this ontology to construct interest-based features and shown that the resulting features improve the performance of various classifiers for predicting friendships in the Live Journal social network. We have also shown that using the interests' ontology, one can address the problem of predicting the interests of Live Journal users, a task that in absence of the ontology is not feasible otherwise as there is an overwhelming number of interests.

The present thesis aims to deduce tasks and characteristics of end users during the open innovation process on the social web. The social web with its communities, forums and blogs affords new prospects as well as unknown challenges for companies, and at the same has increasingly influenced academic research during the last few years. Especially research regarding communication behaviour on the social web as well as social web technologies has currently progressed well. However, in innovation research, social web technologies are currently primarily used to integrate users into the company’s innovation process, for example through company user toolkits or company innovation communities. In those cases users were excluded from their normal social web environment and integrated into a company’s environment, a sort of laboratory environment. Despite this, the present research project will use the natural behaviour, comments and discussions of users within their social web environment to develop and apply a new mixed-method approach with the aim to deduce tasks and characteristics of innovative end users on the social web. To apply the mixed-method approach within a longitudinal case study and to deduce statements and regularities regarding the innovation process on the social web, it was possible to analyse the end user developer online forum of one of the leading open source CRM software technologies. Based on this analysis, the assumptions from an extensive literature analysis could be verified and extended: it could be shown that the expected single innovative user does not exist. In fact, the process from the initial idea to an innovation requires different users with different characteristics and different points of view. They will be deduced, explained and presented within the present thesis.

The UK is required to reduce its greenhouse gas emissions by 80 per cent from 1990 levels, by 2050. Greenhouse gas emissions attributed to the UK higher education sector have increased by 34.5 per cent from 1990 to 2005. Higher education institutions have a unique role in the UK greenhouse gas emissions inventory, beyond management of their own estates and compliance with policy and legislation, higher education institutions have responsibilities as innovators and educators, inspiring students and employees through example and best practice. This study sought to understand acceptability of greenhouse gas emissions reduction policies among employees of a higher education institution. The value-belief-norm theory was used in a questionnaire to understand individual attitudinal factors thought to influence policy acceptability (N=405). Recognising that an employee's attitudinal factors may be influenced by their work colleagues, this study used social network analysis to understand the social context within which individual attitudinal factors sit. Support was found for higher education institutions to reduce their greenhouse gas emissions. Employees found policies that encouraged desired behaviours, such as assistance with train travel costs and working from home, to be more acceptable than policies that discouraged undesired behaviours, such as doubling the price of a car-parking permit. Support was found for the structure and content of the value-belief-norm theory, but logistic regression suggested that it provided a weak explanation of employee policy acceptability, indicating that other factors may have a greater role. Analysis of workplace social networks suggested that employees have small social groups (x̄=8) and do not select to be close to colleagues that reflect their own perspectives. Practitioners and policymakers should seek to address this void in environmental social norms through recruitment of more environmental champions to deliver strong and persuasive pro-environmental messages.

The innovation ecosystem’s online presence continues to grow with the emergence and maturation of ICT-based platforms. With these new channels, a diversity of actors, including firms, scientists, universities, media entities, and individuals, interact to satisfy their information needs and to access and mobilize network-based resources. This research is among a growing number of social science studies examining the advent of social media and its influence on the innovation process, asking, “How do different types of actors use social media to form network linkages, and what kinds of innovative outcomes will result?” To study this complex network activity, I turn to Twitter, the popular microblogging service, and focus on the case of graphene, a novel nanotechnology material consisting of a two-dimensional sheet of carbon atoms. Twitter is one of the world’s most often-used social networks, boasting over 500 million users (200+ million active). Graphene, on the other hand, is a relatively well-bounded area of scientific inquiry with ongoing, concurrent scientific and commercialization activity. The primary sample dataset derives from 34 graphene firms’ friend and followers relationships captured in early 2014. Nine interview transcripts supply qualitative data. The results show that network formation on Twitter is not random and that certain actor relationships predict following linkages. A series of network visualizations show that users agglomerate in communities; these communities exhibit greater density than the larger ecosystem network and a propensity to congeal in topically focused ways. That is, each community indicates a coherent topical focus, suggesting that graphene firms follow specific sets of users in ways that support their information and resource needs. At the micro-level, an unstructured text mining approach to operationalizing and computing information distance shows that increasing amounts of topical distance between any two users decreases the likelihood of a tie existing. Are innovation outcomes more likely to occur in strategically-developed and information-rich social media networks? Drawing on different sources of “behavioral additionality” – or changes in behaviors as a result of social media participation – I identify ex-ante several such plausible outcomes, which could include increased awareness, improved problem solving ability, community development, and greater sales. The qualitative results show that social media participation results in increased awareness of graphene and related ecosystem topics, but engagement is a key tactical maneuver that actors pursue, often in varying ways, to access and mobilize other resources. Policy implications are targeted at intermediary institutions and scientists, while management implications focus on high-technology SMEs. Limitations include alternative theories to explaining social media participation and engagement, methodological issues, and the continuing evolution of social media platforms and usage patterns. Future work is considered to address the temporal nature of network construction and topical growth (or constriction), as well as the ability to map areas of science and technology through social media data.