Maria Carolina Monard

• Abstract

  Open problems in the multi-disciplinary field of Relational Data Mining, which deals with knowledge discovery from relational databases consisting of multiple tables. Understanding the many faces of Receiver Operating Characteristics (ROC) analysis in machine learning. (AU)

  Peter Flash | University of Bristol - Inglaterra, AV.EXT
• Nitin Indurkhya | Nanyang Techonological University - Cingapura, AV.EXT

• Abstract

  Feature Selection plays an important role in machine learning, as irrelevant and/or redundant features may reduce the quality and comprehensibility of the hypotheses induced by supervised learning algorithms. This work proposes the evaluation of a new feature selection strategy using mutual information, which will be estimated using a new method, recently proposed, related to the estimati…

  Evaluating a new feature selection strategy based on mutual information, BP.IC

• Abstract

  The aim of this project is to develop methods and tools for exploratory multi-label data analysis. Using these developed methods, we intend to identify the most relevant characteristics of multi-label data and their influence on existing multi-label classification methods. Visualization tools, as well as several complexity measures of multi-label data, will be implemented in a framework t…

  Exploratory data analysis for multilabel data, BP.IC
• Synthetic Data Generation for Multi-label Learning, BP.IC

(Only some records are available in English at this moment)

• Abstract

  Feature Selection (FS) is an important task in machine learning, which caneffectively reduce the dataset dimensionality by removing irrelevantand/or redundant features. Although a large body of research dealswith FS in single-label data, in which several measures have been proposedto filter out irrelevant features, this is not the case for multi-labeldata where each instance is la…

  Feature selection for multi-label learning, BE.EP.DR

• Abstract

  Due to the increasing number of new applications where examples are annotated with more than one label, multi-label classification has attracted the attention of the academic community. Multi-label classification is being used in an increasing number of applications such as semantic annotation of video and image, functional genomic and music categorization into emotions, to name just a fe…

  Active learning algorithms for multi-label data, BE.EP.DD