Suchergebnisse

Our uses of digital technologies like social media platforms or email leave massive amounts of behavioral traces that are most interesting for social research. Other digital technologies like cell phones allow harnessing behavioral traces for research purposes. Such Digital Behavioral Data consists of genuinely relational records which can be thought of in terms of networks. However, this kind of data requires a shift of perspective from individuals to micro events (e.g., a post on social media) as units of observation and brings established techniques like Social Network Analysis to the center stage. We argue that, using this approach, obtaining individual attributes and attitudes as well as uncovering the micro-macro dynamics of behavior by mining patterns are potentially fruitful applications. We discuss methodological challenges and conclude that social theory is a constitutive pillar for the consolidation of Computational Social Science.

Die Nutzung von digitalen Technologien, wie Social-Media-Plattformen, hinterlässt riesige Mengen an Verhaltensspuren, die für die Sozialforschung überaus interessant sind. Andere digitale Technologien, wie Mobiltelefone, ermöglichen die Erhebung von Verhaltensspuren zu Forschungszwecken. Diese digitalen Verhaltensdaten bestehen aus genuin relationalen Beziehungen, die als Netzwerke betrachtet werden können. Diese Art von Daten erfordert eine Verlagerung der Perspektive von Individuen zu Mikroereignissen (z.B. ein Post in den sozialen Medien) als Beobachtungseinheiten und stellt etablierte Techniken, wie die Analyse sozialer Netzwerke, in den Mittelpunkt. Wir argumentieren, dass die Verwendung dieses Ansatzes, die Ermittlung individueller Eigenschaften und Einstellungen sowie die Offenlegung von Mikro-Makro-Verhaltensdynamiken mittels Auswertung von Mustern potenziell nützliche Anwendungen sind. Wir diskutieren methodische Herausforderungen und gelangen zu der Schlussfolgerung, dass soziale Theorie ein Grundpfeiler für die Konsolidierung der Computational Social Science (computergestützte Sozialwissenschaft) ist.

Rapid Miner Process files and XML test set including the predicted labels for the Linked Data Mining Challenge 2015.

In recent years Linked Open Data (LOD) has matured and gained acceptance across various communities and domains. Large potential of Linked Data technologies is seen for an application in scientific disciplines. In this article, we present use cases and applications for an application of Linked Data in the social sciences. They focus on (a) interlinking domain-specific information, and (b) linking social science data to external LOD sources (e.g. authority data) from other domains. However, several technical and research challenges arise, when applying Linked Data technologies to a scientific domain with its specific data, information needs and use cases. We discuss these challenges and show how they can be addressed. (author's abstract)

Linking Study Descriptions to the Linked Open Data Cloud

At the very beginning of compiling a bibliography, usually only basic information, such as title,
authors and publication date of an item are known. In order to gather additional information about a specific item, one typically has to search the library catalog or use a web search engine.
This look-up procedure implies a manual effort for every single item of a bibliography. In this
technical report we present a proof of concept which utilizes Linked Data technology for the
simple enrichment of sparse metadata sets. This is done by discovering owl:sameAs links between
an initial set of computer science papers and resources from external data sources like DBLP, ACM and the Semantic Web Conference Corpus. In this report, we demonstrate how the link discovery tool Silk is used to detect additional information and to enrich an initial set of records in the computer science domain. The pros and cons of silk as link discovery tool are summarized in the end.

We present a unique collection of four data sets to study social behaviour, collected during international scientific conferences. Interactions between participants were tracked using the SocioPatterns platform, which allows collecting face-to-face physical proximity events every 20 seconds. Through accompanying surveys, we gathered extensive information about the participants: sociodemographic characteristics, Big Five personality traits, DIAMONDS situation perceptions, measure of scientific attractiveness, motivations for attending the conferences, and perceptions of the crowd. Linking the sensor and survey data provides a rich window into social behaviour. At the individual level, the data sets allow personality scientists to investigate individual differences in social behaviour and pinpoint which individual characteristics (e.g., social roles, personality traits, situation perceptions) drive these individual differences. At the group level, the data allow to study the mechanisms responsible for interacting patterns within a scientific crowd during a social, networking and idea-sharing event.

Uncovering how inequality emerges from human interaction is imperative for just societies. Here we show that the way social groups interact in face-to-face situations can enable the emergence of disparities in the visibility of social groups. These disparities translate into members of specific social groups having fewer social ties than the average (i.e., degree inequality). We characterize group degree inequality in sensor-based data sets and present a mechanism that explains these disparities as the result of group mixing and group-size imbalance. We investigate how group sizes affect this inequality, thereby uncovering the critical size and mixing conditions in which a critical minority group emerges. If a minority group is larger than this critical size, it can be a well-connected, cohesive group; if it is smaller, minority cohesion widens inequality. Finally, we expose group under-representation in degree rankings due to mixing dynamics and propose a way to reduce such biases.

Academic search systems aid users in finding information covering specific topics of scientific interest and have evolved from early catalog-based library systems to modern web-scale systems. However, evaluating the performance of the underlying retrieval approaches remains a challenge. An increasing amount of requirements for producing accurate retrieval results have to be considered, e.g., close integration of the system's users. Due to these requirements, small to mid-size academic search systems cannot evaluate their retrieval system in-house. Evaluation infrastructures for shared tasks alleviate this situation. They allow researchers to experiment with retrieval approaches in specific search and recommendation scenarios without building their own infrastructure. In this paper, we elaborate on the benefits and shortcomings of four state-of-the-art evaluation infrastructures on search and recommendation tasks concerning the following requirements: support for online and offline evaluations, domain specificity of shared tasks, and reproducibility of experiments and results. In addition, we introduce an evaluation infrastructure concept design aiming at reducing the shortcomings in shared tasks for search and recommender systems.

The user-study evaluates a vocabulary term recommendation service that is based on how other data providers have used RDF classes and properties in the Linked Open Data cloud. The study compares the machine learning technique Learning to Rank (L2R), the classical data mining approach Association Rule mining (AR), and a baseline that does not provide any recommendations. This data collection comprises the raw results of this user-study in SPSS format.

The main contribution of the survey is to condense and aggregate the experience and knowledge of Linked Data experts and practitioners regarding the most common strategies to reuse vocabularies when modeling Linked Open Data in a real-world scenario.

We present a longitudinal data set of four scientific conferences during which we have collected face-to-face contacts along with extensive information about the participants (see https://doi.org/10.7802/2352 ). Interactions between participants were monitored using the SocioPatterns platform, which allows to detect and record face-to-face physical proximity events every 20 seconds in a well-defined social context. This data set contains the contact data, i.e. the interaction events that were collected during the four studies. Participants were not linked through the four studies.

We present a longitudinal data set of four scientific conferences during which we have collected face-to-face contacts along with extensive information about the
participants. Interactions between participants were monitored using the SocioPatterns platform, which allows to detect and record face-to-face physical proximity events every 20 seconds in a well-defined social context (see https://doi.org/10.7802/2351 ). This data set contains socio-demographic information about the participants, collected via a survey. Participants were not linked through the four studies.