NESG

Icono Icono

Icono Icono

Data processing and networkmetrics

Members

Description

The networkmetrics research line seeks to take advantage of multivariate analysis and machine learning tools to tackle problems in communication networks, with cybersecurity as main example. An effective detection of cybersecurity incidents requires the combination of several and disparate data sources. This makes cybersec a typical Big Data problem, where the challenge is to handle tons of information from heterogeneous sources at a fastpace. In NESG, we develop new analysis methods to handle Multivariate Big Data, which are also of value in applications like IoT monitoring or Industry 4.0, and in other domains, like chemometrics, bioinformatics and personalized medicine.

  

Publications

  • Camacho, J., Acar, E., Rasmussen, M. & Bro, R. (2019). X-CAN: Cross-Penalized Component Analysis. Scandinavian Symposium on Chemometrics (SSC16), . [More] 
  • Camacho, J., Smilde, A. K., Saccenti, E. & Westerhuis, J. A. (2019). All sparse PCA models are wrong, but some are useful. Scandinavian Symposium on Chemometrics (SSC16), . [More] 
  • Camacho, J., Smilde, A. K., Saccenti, E. & Westerhuis, J. A. (2019). All Sparse PCA Models Are Wrong, But Some Are Useful. Part I: Computation of Scores, Residuals and Explained Variance. arXiv Preprint. Retrieved from https://arxiv.org/abs/1907.03989. [More] 
  • Camacho, J., Acar, E., Rasmussen, A. & Bro, R. (2019). Cross-product Penalized Component Analysis (XCAN). arXiv preprint. Retrieved from https://arxiv.org/abs/1907.00032. [More] 
  • Camacho, J., Therón, R., García-Giménez, J. M., Maciá-Fernández, G. & García-Teodoro, P. (2019). Group-Wise Principal Component Analysis for Exploratory Intrusion Detection. IEEE Access, 7, 113081-113093. [More]