2011
DOI: 10.1590/s0103-17592011000500002
|View full text |Cite
|
Sign up to set email alerts
|

Aprendizado supervisionado com conjuntos de dados desbalanceados

Abstract: Supervised Learning with Imbalanced Data Sets: An OverviewTraditional learning algorithms induced by complex and highly imbalanced training sets may have difficulty in distinguishing between examples of the groups. The tendency is to create classification models that are biased toward the overrepresented (majority) class, resulting in a low rate of recognition for the minority group. This paper provides a survey of this problem which has attracted the interest of many researchers in recent years. In the scope … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
2
0
7

Year Published

2013
2013
2023
2023

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 9 publications
(9 citation statements)
references
References 78 publications
0
2
0
7
Order By: Relevance
“…We evaluated the predictive capability of the mechanistic models and the classification algorithms using widespread performance metrics for binary classification problems. According to Castro and Braga (2011), these criteria either focus on the detection of the minority class in unbalanced classification problems or consider the discrimination of both classes as having the same relevance. All metrics used in this assessment yield values between 0 (poor performance) and 1 (high performance).…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…We evaluated the predictive capability of the mechanistic models and the classification algorithms using widespread performance metrics for binary classification problems. According to Castro and Braga (2011), these criteria either focus on the detection of the minority class in unbalanced classification problems or consider the discrimination of both classes as having the same relevance. All metrics used in this assessment yield values between 0 (poor performance) and 1 (high performance).…”
Section: Resultsmentioning
confidence: 99%
“…It is calculated as the harmonic mean of the recall and precision, usually being β = 1 (12). Β is used to determine the relative importance of recall and precision (Castro and Braga, 2011 …”
Section: Metrics For Performance Evaluationmentioning
confidence: 99%
“…Muitos trabalhos na área de reconhecimento de padrões tem a análise de desempenho de seus algoritmos baseada na acurácia dos seus classificadores. Entretanto, sabe-se que a acurácia pode mascarar taxas de erros de classificação quando se trabalha com classes desbalanceadas [21].…”
Section: Critério De Desempenho -áRea Abaixo Da Curva Rocunclassified
“…Os gráficos para as curvas ROC são bidimensionais, nos quais o eixo das ordenadas plota-se a sensibilidade e no eixo das abscissas a especificidade [22]. A curva ROC de um classificador ideal possui o formato da função Heaviside (Heaviside step function) [21].…”
Section: Critério De Desempenho -áRea Abaixo Da Curva Rocunclassified
“…En el caso específico de dos clases, este problema se identifica porque existe un número muy pequeño de instancias de una de las clases, en comparación con el número de instancias de la otra. En la literatura existen diferentes estudios que abordan el problema del desbalanceo entre clases, muchos proponen soluciones específicas al problema [13]- [15] y otros pocos estudian las causas del mismo [16]- [19]. Sin embargo, la conclusión general es que ante un conjunto de datos de entrenamiento desbalanceado, los algoritmos de aprendizaje tradicionales generan superficies de decisión que tienden a estar sesgadas por la clase mayoritaria, como se ilustra en la Fig.…”
Section: El Problema De Clases Desbalanceadasunclassified