Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

O2, repositorio institucional de la UOC
Publication Artículo científico (JournalArticle)

Utility-preserving privacy protection of nominal data sets via semantic rank swapping

  • Rodríguez García, Mercedes
  • Batet, Montserrat
  • Sánchez, David
Personal data are of great interest for research but, at the same time, they pose a serious privacy risk. Therefore, appropriate data protection measures should be undertaken by the data controller before making personal data available for secondary use. Also, such data protection should be done in a way that data are still useful for analysis. In the last years, a plethora of data protection mechanisms have been proposed. Among them, rank swapping is considered one of the best with respect to disclosure risk minimization and data utility preservation. Because rank swapping is based on sorting input data to swap values that are close to each other, in principle, it is a method restricted to numerical and ordinal categorical data. However, a significant amount of personal data currently compiled and used in data analysis are nominal, and their utility depends on the semantics they convey. To properly cope with this type of data, in this paper, we present rank swapping methods capable of protecting nominal data from a semantic perspective. Specifically, by exploiting ontologies, our methods are able to protect nominal data while properly preserving their semantics and, thus, their analytical utility. For that, we provide a suitable binary relation to semantically sort nominal data. Our proposal is capable of managing both independent individual attributes and non-independent multivariate data sets, being the latter especially relevant for data analysis. Empirical experiments carried on real clinical records and using a standard medical ontology show that our methods are able to preserve the semantic features of nominal data significantly better than standard permutation mechanisms.
O2, repositorio institucional de la UOC
Publication Artículo científico (JournalArticle)

Leveraging synonymy and polysemy to improve semantic similarity assessments based on intrinsic information content

  • Batet, Montserrat
  • Sanchez, David
Semantic similarity measures based on the estimation of the information content (IC) of concepts are currently regarded as the state of the art. Calculating the IC in an intrinsic (i.e., ontology-based) way is particularly convenient due to its accuracy and lack of dependency on annotated corpora. Intrinsic IC calculation models estimate concept probabilities from the taxonomic knowledge (i.e., number of hyponyms and/or hypernyms of the concepts) modelled in an ontology. In this paper, we aim to improve the intrinsic calculation of the IC by leveraging not only the hyponyms and hypernyms of concepts, but also the explicit evidences of synonymy and polysemy that ontologies such as WordNet also model. Specifically, we propose a more accurate intrinsic estimation of the concepts’ probabilities in which the IC calculation relies. We evaluate the accuracy of our proposal through a set of comprehensive experiments in which our IC calculation model is tested on a variety of IC-based similarity measures and benchmarks. Experimental results show that our proposal obtains consistently good accuracies, which vary less across measures and benchmarks than the most prominent intrinsic IC calculation models available in the literature.