We study the attention dynamics towards prominent scientists by analyzing different Web signals. In our research we demonstrate how success of the scientist is related to the success of the scientific field he is working in. We use unsupervised learning techniques to discover the attention patterns in the time series. Based on these patterns we try to predict the future success of the scientist.
We investigate the relationship between semantics for formal argumentation and measures from social networking theory. In particular, we consider using matrix exponentials, which are measures used for link prediction and recommendation in social networks, as a way to measure acceptability of arguments in abstract argumentation frameworks.
Word embeddings are mappings of words to dense, real-valued vectors. The word embeddings that were created as a by-product of training neural networks for the task of language modelling were shown to encoded certain syntactic and semantic regularities. This talk will give a quick introduction to neural networks, will explain how they can be applied to the task of language modelling, and will discuss semantic properties of word embeddings. Lastly, the goal of Lukas' thesis will be presented.
Heuristic evaluation is an informal method of usability analysis. In this presentation, a set of usability heuristics for evaluating the usability of an eye controlled interfaces has been designed .
Since an RDF store may contain data from different sources with varying levels of trustworthiness, it is important to store provenance information about the data and allow the users to request these information.
For this reason, the aim of this presentation is to present the progress of the RDF and SPARQL extension developed in my bachelor thesis.
The RDF and SPARQL extension allows storing and querying of provenance information in an RDF store.
In this talk, I will present multiple analysis methods that can be performed on network dataset collections which contain a large number of different datasets. Although network datasets (both social and otherwise) are used in a large fraction of studies in data mining and other areas, many papers base their work on only a single, or a small number of datasets.
In this talk, I will discuss the methods for empirical experiment (user studies) in human-computer interaction research. This is an introductory talk if you are keen on conducting an experiment to evaluate a new computer interface or interaction technique.
Microtask crowdsourcing has become an extremely useful method to solve problems that are difficult for machines. However, there are still many challenges to overcome. One way to identify the problems that should be tackled is to pay attention to what the users of crowdsourcing have to say. In this thesis, I studied messages posted on Twitter and forums, that referred to four major microtask crowdsourcing platforms. I analysed who posted the messages, the step of the crowdsourcing process they refer to, the purpose of the messages and their sentiment. This is a first step toward the identification of strengths and weaknesses of crowdsourcing platforms.
As a free and collaborative knowledge base, created and maintained by volunteers, Wikidata may represent knowledge from different points of view. These points of view can be conflicting, especially when the data comes from different information sources.
Finding suitable datasets is a major task within many science topics. Especially social sciences rely heavily on existing studies since the creation of new data with hundreds of participants from different countries over a longer periods of time is resource intensive, so the reuse and recombination of existing studies is crucial.