Istituto di Scienza e Tecnologie dell'Informazione     
Liu R., Wang W. H., Monreale A., Pedreschi D., Giannotti F., Wenge G. AUDIO: an integrity auditing framework of outlier-mining-as-a-service systems. In: ECML PKDD 2012 - Machine Learning and Knowledge Discovery in Databases European Conference (Bristol, UK, 24-28 September 2012). Proceedings, vol. II pp. 1 - 18. Peter A. Flach , Tijl De Bie, Nello Cristianini (eds.). (Lecture Notes in Computer Science, vol. 7524). Springer, 2012.
Spurred by developments such as cloud computing, there has been considerable recent interest in the data-mining-as-a-service paradigm. Users lacking in expertise or computational resources can outsource their data and mining needs to a third-party service provider (server). Outsourcing, however, raises issues about result integrity: how can the data owner verify that the mining results returned by the server are correct? In this paper, we present AUDIO, an integrity auditing framework for the specific task of distance-based outlier mining outsourcing. It provides efficient and practical verification approaches to check both completeness and correctness of the mining results. The key idea of our approach is to insert a small amount of artificial tuples into the outsourced data; the artificial tuples will produce artificial outliers and non-outliers that do not exist in the original dataset. The server's answer is verified by analyzing the presence of artificial outliers/non-outliers, obtaining a probabilistic guarantee of correctness and completeness of the mining result. Our empirical results show the effectiveness and efficiency of our method.
URL: http://www.springerlink.com/content/h4811066nl6x8227/
DOI: 10.1007/978-3-642-33486-3_1
Subject Integrity
Outlier detection
E.3 Data Encryption

Icona documento 1) Download Document PDF
Icona documento 2) Download Document PDF

Icona documento Open access Icona documento Restricted Icona documento Private


Per ulteriori informazioni, contattare: Librarian http://puma.isti.cnr.it

Valid HTML 4.0 Transitional