Auswahl der wissenschaftlichen Literatur zum Thema „Data mining – social aspects“

Analyzing a huge amount of data becomes a tricky challenge and an opportunity for data miners and businessmen today. Knowledge management processes can deal with big knowledge source to find tacit intelligence making businesses more agile and effective. Data mining is a powerful tool working with big data to create capabilities of forecasting and analysis. Yet there is a lack of research on where and how data mining can add value in knowledge management processes in organizations to maximize valuable knowledge for innovation and business management. The knowledge management processes of a psychiatry section in a Swedish hospital was used as a case study for this thesis. Interviews with manager, psychiatrist, auxiliary nurse and data scientists are conducted. Collected data is analyzed to create values of data mining based on a value creation framework through the knowledge management processes of psychiatry section in the hospital. Relying on this process, the limitations and strengths are exposed; whereby, a data mining implementing framework is formulated, and potentials of data mining for the process are suggested to support for all employees of psychiatry section in the hospital in decision making and caring for patients.

The availability and scale of user generated data in online systems raises tremendous challenges and opportunities to analytic study of human activities. Effective modeling of online human activities is not only fundamental to the understanding of human behavior, but also important to the online industry. This thesis focuses on developing models and algorithms to predict human activities in online systems and to improve the algorithmic design of personalized/socialized systems (e.g., recommendation, advertising, Web search systems). We are particularly interested in three types of online user activities, i.e., decision making, social interactions and user-generated contents. Centered around these activities, the thesis focuses on three challenging topics: 1. Behavior prediction, i.e., predicting users' online decisions. We present Collaborative-Competitive Filtering, a novel game-theoretic framework for predicting users' online decision making behavior and leverage the knowledge to optimize the design of online systems (e.g., recommendation systems) in respect of certain strategic goals (e.g., sales revenue, consumption diversity). 2. Social contagion, i.e., modeling the interplay between social interactions and individual behavior of decision making. We establish the joint Friendship-Interest Propagation model and the Behavior-Relation Interplay model, a series of statistical approaches to characterize the behavior of individual user's decision making, the interactions among socially connected users, and the interplay between these two activities. These techniques are demonstrated by applications to social behavior targeting. 3. Content mining, i.e., understanding user generated contents. We propose the Topic-Adapted Latent Dirichlet Allocation model, a probabilistic model for identifying a user's hidden cognitive aspects (e.g., knowledgability) from the texts created by the user. The model is successfully applied to address the challenge of ``language gap" in medical information retrieval.

Businesses today are often operating in a highly competitive environment where information is a noticeably valuable asset. Businesses are therefore in need of powerful tools for extracting actionable business knowledge. Research show that SME companies are lagging behind large companies in the use of data analytics; even though they know the potential benefits. We want to study and compare different tools for data analytics and how they can be used by small companies. Our research questions are therefore: what analytical tools are today available on the market, and what are their possibilities and challenges for small companies? And: how can these analytical tools aid in the development of a business, product or service? We conclude in our research that there are several data analytics tools available for small businesses, that their different usages can be applied successfully and without big cost, and that their relevance, both in business development and innovation, depends on the business objectives and goals of their utilization.

<p> Powered by big data infrastructures, social network platforms are gathering data on many aspects of our daily lives. The online social world is reflecting our physical world in an increasingly detailed way by collecting people's individual biographies and their various of relationships with other people. Although massive amount of social data has been gathered, an urgent challenge remain unsolved, which is to discover meaningful knowledge that can empower the social platforms to really understand their users from different perspectives. </p><p> Motivated by this trend, my research addresses the reasoning and mathematical modeling behind interesting phenomena on social networks. Proposing graph based data mining framework regarding to heterogeneous data sources is the major goal of my research. The algorithms, by design, utilize graph structure with heterogeneous link and node features to creatively represent social networks' basic structures and phenomena on top of them.</p><p> The graph based heterogeneous mining methodology is proved to be effective on a series of knowledge discovery topics, including network structure and macro social pattern mining such as magnet community detection (87), social influence propagation and social similarity mining (85), and spam detection (86). The future work is to consider dynamic relation on social data mining and how graph based approaches adapt from the new situations. </p>

Social media services have a growing impact in our society. Individuals often rely on social media to get their news, decide which products to buy or to communicate with their friends. As consequence of the widespread adoption of social media, a large volume of data on how users behave is created every day and stored into large databases. Learning how to analyze and extract useful knowledge from this data has a number of potential applications. For instance, a deeper understanding on how legitimate users interact with social media services could be explored to design more accurate spam and fraud detection methods. This PhD research is based on the following hypothesis: data generated by social media users present patterns that can be exploited to improve the effectiveness of tasks such as prediction, forecasting and modeling in the domain of social media. To validate our hypothesis, we focus on designing data mining methods tailored to social media data. The main contributions of this PhD can be divided into three parts. First, we propose Act-M, a mathematical model that describes the timing of users actions. We also show that Act-M can be used to automatically detect bots among social media users based only on the timing (i.e. time-stamp) data. Our second contribution is VnC (Vote-and-Comment), a model that explains how the volume of different types of user interactions evolve over time when a piece of content is submitted to a social media service. In addition to accurately matching real data, VnC is useful, as it can be employed to forecast the number of interactions received by social media content. Finally, our third contribution is the MFS-Map method. MFS-Map automatically provides textual annotations to social media images by efficiently combining visual and metadata features. Our contributions were validated using real data from several social media services. Our experiments show that the Act-M and VnC models provided a more accurate fit to the data than existing models for communication dynamics and information diffusion, respectively. MFS-Map obtained both superior precision and faster speed when compared to other widely employed image annotation methods.<br>O impacto dos serviços de mídia social em nossa sociedade é crescente. Indivíduos frequentemente utilizam mídias sociais para obter notícias, decidir quais os produtos comprar ou para se comunicar com amigos. Como consequência da adoção generalizada de mídias sociais, um grande volume de dados sobre como os usuários se comportam é gerado diariamente e armazenado em grandes bancos de dados. Aprender a analisar e extrair conhecimentos úteis a partir destes dados tem uma série de potenciais aplicações. Por exemplo, um entendimento mais detalhado sobre como usuários legítimos interagem com serviços de mídia social poderia ser explorado para projetar métodos mais precisos de detecção de spam e fraude. Esta pesquisa de doutorado baseia-se na seguinte hipótese: dados gerados por usuários de mídia social apresentam padrões que podem ser explorados para melhorar a eficácia de tarefas como previsão e modelagem no domínio das mídias sociais. Para validar esta hipótese, foram projetados métodos de mineração de dados adaptados aos dados de mídia social. As principais contribuições desta pesquisa de doutorado podem ser divididas em três partes. Primeiro, foi desenvolvido o Act-M, um modelo matemático que descreve o tempo das ações dos usuários. O autor demonstrou que o Act-M pode ser usado para detectar automaticamente bots entre usuários de mídia social com base apenas nos dados de tempo. A segunda contribuição desta tese é o VnC (Vote-and- Comment), um modelo que explica como o volume de diferentes tipos de interações de usuário evolui ao longo do tempo quando um conteúdo é submetido a um serviço de mídia social. Além de descrever precisamente os dados reais, o VnC é útil, pois pode ser empregado para prever o número de interações recebidas por determinado conteúdo de mídia social. Por fim, nossa terceira contribuição é o método MFS-Map. O MFS-Map fornece automaticamente anotações textuais para imagens de mídias sociais, combinando eficientemente características visuais e de metadados das imagens. As contribuições deste doutorado foram validadas utilizando dados reais de diversos serviços de mídia social. Os experimentos mostraram que os modelos Act-M e VnC forneceram um ajuste mais preciso aos dados quando comparados, respectivamente, a modelos existentes para dinâmica de comunicação e difusão de informação. O MFS-Map obteve precisão superior e tempo de execução reduzido quando comparado com outros métodos amplamente utilizados para anotação de imagens.

Social Networking Sites (SNS) are ubiquitous within modern society, forming communications networks that span across cultural and geographical boundaries. The information posted to these sites provide useful insights into individuals, but can also provide a wealth of information that can be used for further analysis into the surrounding environment. Three main challenges limit the use of this information in applications: the quantity of data is often unmanageable, there is a significant amount of data unavailable for use due to a lack of generic interfaces for access, and there is difficulty in integrating multiple disparate social data sources. The overall aim of the research described in this thesis is to advance the field of data science and improve accessibility of social data in analytical applications, in both academic and commercial settings. This aim has been addressed with three primary contributions; new algorithms to efficiently locate and collect relevant social data, new methods of performing unsupervised data extraction from generic social sites, and the development and subsequent empirical evaluation of a framework to facilitate the collection, integration, storage and presentation of social data for use in applications. The first contribution was the presentation of a search query optimisation algorithm designed to reduce the amount of noise resulting from social data collection by learning from collected content and iteratively building new query keyword sets. The algorithm was empirically evaluated and the results indicated that it provides significantly more data than existing search tools while minimising signal-to-noise ratio. The second contribution aimed to improve access to social data available on Web 2.0 sites but without any existing interface access to the data. The algorithm is designed to extract social data from sites without any a priori knowledge of design or page layout. Its efficacy was empirically evaluated against a testbed consisting of popular news and current affairs websites. Results indicated that the algorithm was very effective at unsupervised retrieval of social data. The third major contribution presented a framework that integrated the previous two contributions into a framework designed to streamline use of social data in academic and commercial applications. The generic, component-based design was evaluated in real-world scenarios and determined to provide a full social collection and analytics workflow in an extensible and scalable manner. This research has theoretical and practical implications for the use of social data in analytical research and commercial use. It extends the data extraction field to include user-generated content, while providing new avenues for performing semi-intelligent social data sourcing, and significantly improves the accessibility of social data.

This thesis improves the process of recommending people to people in social networks using new clustering algorithms and ranking methods. The proposed system and methods are evaluated on the data collected from a real life social network. The empirical analysis of this research confirms that the proposed system and methods achieved improvements in the accuracy and efficiency of matching and recommending people, and overcome some of the problems that social matching systems usually suffer.