Methods and algorithms in unsupervised and semi-supervised machine learning

• Grant number: 13/18698-4
• Support type: Regular Research Grants
• Duration: February 01, 2014 - January 31, 2016
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Ricardo José Gabrielli Barreto Campello
• Grantee: Ricardo José Gabrielli Barreto Campello
• Home Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

This document formalizes a research proposal for funding from the Research Foundation of the State of São Paulo (FAPESP). The main objective is to support the research group of the applicant for a period of two years. The group is mainly constituted by the researcher and his students at ICMC / USP, but the project will also involve international collaborations with the University of Alberta in Canada and the Ludwig-Maximilians-Universität in Germany. With regard to research topics, it is an "umbrella-like"project which comprises subprojects to be developed primarily by MSc and PhD students, directly under the supervision of the applicant in Brazil or through collaborations abroad. In general, the proposal aims to investigate methods of machine learning for data mining problems. The areas of research focus mainly on the paradigms of unsupervised and semi-supervised learning, with particular focus on the areas of cluster analysis and anomaly (or outlier) detection. In the context of data clustering, research topics of interest include, amongothers, hierarchical clustering, density-based clustering, subspace clustering, semi-supervised clustering, parallel and distributed clustering, overlapping clustering, bi-clustering and cluster validation. In the context of outlier detection, topics of interest include outlier detection ensembles, validation of unsupervised outlier detection, and computationally scalable methods, among others. With regard to application areas, the project will focus on general purpose mathematical and computational tools, but it may also involve specific domains such as gene expression data analysis. (AU)

Scientific publications

(5)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

PADILHA, VICTOR A.; CAMPELLO, RICARDO J. G. B. A systematic comparative evaluation of biclustering techniques. BMC Bioinformatics, v. 18, JAN 23 2017. Web of Science Citations: 23.

CAMPOS, GUILHERME O.; ZIMEK, ARTHUR; SANDER, JORG; CAMPELLO, RICARDO J. G. B.; MICENKOVA, BARBORA; SCHUBERT, ERICH; ASSENT, IRA; HOULE, MICHAEL E. On the evaluation of unsupervised outlier detection: measures, datasets, and an empirical study. DATA MINING AND KNOWLEDGE DISCOVERY, v. 30, n. 4, p. 891-927, JUL 2016. Web of Science Citations: 71.

HORTA, DANILO; CAMPELLO, RICARDO J. G. B. Comparing Hard and Overlapping Clusterings. JOURNAL OF MACHINE LEARNING RESEARCH, v. 16, p. 2949-2997, DEC 2015. Web of Science Citations: 3.

NALDI, M. C.; CAMPELLO, R. J. G. B. Comparison of distributed evolutionary k-means clustering algorithms. Neurocomputing, v. 163, n. SI, p. 78-93, SEP 2 2015. Web of Science Citations: 17.

CAMPELLO, RICARDO J. G. B.; MOULAVI, DAVOUD; ZIMEK, ARTHUR; SANDER, JOERG. Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection. ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, v. 10, n. 1 JUL 2015. Web of Science Citations: 70.