Metrics for Multiset-Theoretic Subgraphs

An information theoretic approach for privacy metrics

Organizations often need to release microdata without revealing sensitive information. To this scope, data are anonymized and, to assess the quality of the process, various privacy metrics have been proposed, such as k-anonymity, `-diversity, and t-closeness. These metrics are able to capture different aspects of the disclosure risk, imposing minimal requirements on the association of an indivi...

Information-theoretic metrics for security and privacy

In this thesis, we study problems in cryptography, privacy and estimation through the information-theoretic lens. We introduce information-theoretic metrics and associated results that shed light on the fundamental limits of what can be learned from noisy data. These metrics and results, in turn, are used to evaluate and design both symmetric-key encryption schemes and privacy-assuring mappings...

Object recognition via information-theoretic measures/metrics

Using Subgroup Discovery Metrics to Mine Interesting Subgraphs

While extensive work has been done in both graph mining and subgroup discovery, the potential benefits of combining the two fields have not been well studied. We propose, implement, and evaluate an adaption of an existing subgroup discovery algorithm to mine graph data. Our experiments use two different metrics from the subgroup discovery literature to demonstrate value in using such metrics to...

User-Specific Decision-Theoretic Accuracy Metrics for Collaborative Filtering

Accuracy is a fundamental dimension for the effectiveness of recommender systems. Several accuracy metrics have been investigated in the literature. However, we argue, these metrics are not sufficiently user-specific. In previous work, we proposed accuracy metrics that take into account a userspecific pointwise decision threshold. In this paper, we present even more user-specific accuracy metri...