Overview

Understanding patterns in complex, high-dimensional data (often totally unlabeled or only partially labeled) has been a key ingredient for knowledge discovery in different domains. In particular, the analysis of large data sets involves a myriad of theoretical, algorithmic and computational challenges that require an integrated effort of inter-disciplinary areas such as data mining, machine learning, complex systems, multi-variate analysis, statistics , databases, distributed computing systems, among others.

Our research aim to meet the growing demand for modern, inter-disciplinary approaches to Data Science, Big Data analytics and Knowledge Discovery from Databases (KDD). Our research projects have theoretical or applied emphasis in these areas, with special focus on (but not restricted to) semi-supervised and unsupervised approaches for pattern analysis / modeling of structured, unstructured, or mixed data.

Research Topics

  • Anomaly / Outlier Detection
  • Active, Inductive and Transductive Learning
  • Bioinformatics
  • Complex Networks
  • Data Cluster Analysis (clustering)
  • Data Mining: unsupervised, semi-supervised and supervised
  • Data Mining: structured, unstructured and mixed data
  • Data Streams Mining
  • Data Preparation: imputation, feature selection, dimensionality reduction, etc.
  • Ensembles: classifiers, clusterings and outlier detectors
  • Graph and Text Mining
  • Machine Learning: unsupervised, semi-supervised and supervised
  • Nonlinear Regression
  • One Class Learning
  • Parallel and Distributed Data Mining
  • Pattern Classification
  • Recommender Systems
  • Social Network Analysis
  • Statistical Learning: Bayesian methods
  • Transfer Learning