Andre Gustavo Maletzke

Andre Gustavo Maletzke

Auto-generated profile, based on publicly available information

Andre Gustavo Maletzke

Universidade de São Paulo, São Carlos, Brazil

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 July 2020 in Data Mining and Knowledge Discovery
Verified
Journal Article
Closed access
PDF file
VSVinicius M. A. Souza
Vinicius M. A. Souza
DRDenis M. dos Reis
Denis M. dos Reis
AMAndré G. Maletzke
André G. Maletzke
GBGustavo E. A. P. A. Batista
Gustavo E. A. P. A. Batista

1 November 2016 in Knowledge-Based Systems
Verified
Journal Article
Closed access
WZWillian Zalewski
Willian Zalewski
FSFabiano Silva
Fabiano Silva
AMA.G. Maletzke
A.G. Maletzke
CFC.A. Ferrero
C.A. Ferrero

17 July 2019 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
AMAndré Maletzke
André Maletzke
DRDenis Dos Reis
Denis Dos Reis
ECEverton Cherman
Everton Cherman
GBGustavo Batista
Gustavo Batista
Quantification is an expanding research topic in Machine Learning literature. While in classification we are interested in obtaining the class of individual observations, in quantification we want to estimate the total number of instances that belong to each class. This subtle difference allows the development of several algorithms that incur smaller and more consistent errors than counting the classes issued by a classifier. Among such new quantification methods, one particular family stands out due to its accuracy, simplicity, and ability to operate with imbalanced training samples: Mixture Models (MM). Despite these desirable traits, MM, as a class of algorithms, lacks a more in-depth understanding concerning the influence of internal parameters on its performance. In this paper, we generalize MM with a base framework called DyS: Distribution y-Similarity. With this framework, we perform a thorough evaluation of the most critical design decisions of MM models. For instance, we assess 15 dissimilarity functions to compare histograms with varying numbers of bins from 2 to 110 and, for the first time, make a connection between quantification accuracy and test sample size, with experiments covering 24 public benchmark datasets. We conclude that, when tuned, Topsøe is the histogram distance function that consistently leads to smaller quantification errors and, therefore, is recommended to general use, notwithstanding Hellinger Distance’s popularity. To rid MM models of the dependency on a choice for the number of histogram bins, we introduce two dissimilarity functions that can operate directly on observations. We show that SORD, one of such measures, presents performance that is slightly inferior to the tuned Topsøe, while not requiring the sensible parameterization of the number of bins.

12 October 2018 in Journal of the Brazilian Computer Society
Verified
Journal Article
Open access
AMAndré G. Maletzke
André G. Maletzke
DRDenis M. dos Reis
Denis M. dos Reis
GBGustavo E. A. P. A. Batista
Gustavo E. A. P. A. Batista
In the last years, learning from data streams has attracted the attention of researchers and practitioners due to its large number of applications. These applications have motivated the research community to propose a significant amount of methods to solve problems in diverse tasks, more prominently in classification, clustering, and anomaly detection. However, a relevant task known as quantification has remained mostly unexplored. The quantification goal is to provide an estimate of the class prevalence in an unlabeled set. Recently, we proposed the SQSI algorithm to quantify data streams with concept drifts. SQSI uses a statistical test to identify concept drifts and retrain the classifiers. However, the retraining involves requiring the labels for all newly arrived instances. In this paper, we extend SQSI algorithm by exploring instance selection techniques allied to semi-supervised learning. The idea is to request the classes of a smaller subset of recent examples. Our experiments demonstrate that SQSI’s extension significantly reduces the dependency on actual labels while maintaining or improving the quantification accuracy.

1 October 2017
Verified
Conference Paper
Closed access
AMAndre G. Maletzke
Andre G. Maletzke
DRDenis M. dos Reis
Denis M. dos Reis
GBGustavo E.A.P.A. Batista
Gustavo E.A.P.A. Batista
In the last decades, learning from data streams has attracted the attention of researchers and practitioners due to its large number of applications. These applications have motivated the research community to propose a significant number of methods that can be used to solve problems in diverse tasks, more prominently in classification, prediction, and clustering. However, a relevant task known as quantification has remained largely unexplored. The quantification goal is to provide an estimate of the class prevalence in an unlabeled set. In this paper, we discuss the relevance and challenges of quantification for data streams and compare how it differs from the batch setting, in which quantification has attracted more attention from the research community. We propose an algorithm to estimate the class distribution in a data stream and frame our algorithm in the active learning framework. In addition, we define two other approaches as baseline and topline strategies for this problem. The experimental results demonstrate that our algorithm has significantly higher quantification accuracy than the baseline and almost as large as the topline while requiring a fraction of the true labels requested by the latter approach.

19 July 2018
Verified
Conference Paper
Open access
DRDenis Moreira dos Reis
Denis Moreira dos Reis
AMAndré Maletzke
André Maletzke
DSDiego F. Silva
Diego F. Silva
GBGustavo E. A. P. A. Batista
Gustavo E. A. P. A. Batista
Many real-world applications in the batch and data stream settings with data shift pose restrictions to the access to class labels after the deployment of a classification or quantification model. However, a significant portion of the data stream literature assumes that actual labels are instantaneously available after issuing their corresponding classifications. In this paper, we explore a different set of assumptions without relying on the availability of class labels. We assume that, although the distribution of the data may change over time, it will switch between one of a handful of well-known distributions. Still, we allow the proportions of the classes to vary. In these conditions, we propose the first method that can accurately identify the correct context of data samples and simultaneously estimate the proportion of the positive class. This estimate can be further used to adjust a classification decision threshold and improve classification accuracy. Finally, the method is very efficient regarding time and memory requirements, fitting data stream applications.