Semantische Suche

In this thesis, we aim to offer location-based services to users of mOSNs while guaranteeing that an adversary, including the service provider, will not be able to learn the locations of the users (location secrecy) and the relationship existing between them (relationship secrecy). We consider both linking attacks and collusion attacks. We propose two approaches R-mobishare and V-mobishare, which combine existing cryptographic techniques. Both approaches use, among others, private broadcast encryption and homomorphic encryption. Private broadcast encryption is used to protect the relationships existing between users, and homomorphic encryption is used to protect the location of the users. Our system allows users to query their nearby friends. Next, we prove that our proposed approaches fulfill our secrecy guarantees, i.e., location and relationship secrecy. Finally, we evaluate the query performance of our proposed approaches and use real online social networks to compare their performance. The result of our experiments shows that in a region with low population density such as suburbs, our first approach, R-mobishare, performs better than our approach V-mobishare. On the contrary, in a region with high population density such as downtown, our second approach, V-mobishare, perform better than R-mobishare.

Um diese automatisiert finden und beheben oder ihnen bestenfalls sogar vorbeugen zu können, bedarf es der Verknüpfung von Text- und Modellentitäten. Dieses Problem wurde in der vorzustellenden Arbeit angegangen. Dabei wurden Wortvektoren verwendet, um Ähnlichkeiten zwischen Wörtern finden zu können.

However, modern systems typically need to solve multiple problems simultaneously. E.g. in order to find high correlations in a dataset, one needs to examine each pair of variables. This is challenging, in particular if the number of variables is large and the data evolves dynamically.

This thesis focuses on the following question: How should one distribute resources at anytime, in order to maximize the overall quality of multiple targets? First, we define the problem, considering various notions of quality and user requirements. Second, we propose a set of strategies to tackle this problem. Finally, we evaluate our strategies via extensive experiments.

While "high-dimensionality" and the "streaming setting" provide two unique sets of challenges, we observe that the existing mining algorithms only address them separately. Thus, our plan is to propose a novel algorithm, which keeps track of the subspaces of interest in high-dimensional data streams over time. We quantify the relevance of subspaces via a so-called "contrast" measure, which we are able to maintain incrementally in an efficient way. Furthermore, we propose a set of heuristics to adapt the search for the relevant subspaces as the data and the underlying distribution evolves.

We show that our approach is beneficial as a feature selection method and as such can be applied to extend a range of knowledge discovery tasks, e.g., "outlier detection", in high-dimensional data-streams.