Under-sampling by algorithm with performance guaranteed for class-imbalance problem

Jindaluang, Wattana; Chouvatut, Varin; Kantabutra, Sanpawat

doi:10.1109/icsec.2014.6978197

Cited by 12 publications

(3 citation statements)

References 7 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Under-sampling methods, samples from the majority class are discarded until the number of samples in each class are nearly equal while preserving valuable information for learning [25,26]. However, it is inevitable that when under-sampling the dataset, some s amp les t h at are meaningful to the training model may be ignored [27,28]. After all, different under-sampling methods have different filtering principles.…”

Section: ) Data-level Methodsmentioning

confidence: 99%

A Comparative Performance Analysis of Data Resampling Methods on Imbalance Medical Data

et al. 2021

View full text Add to dashboard Cite

Medical datasets are usually imbalanced, where negative cases severely outnumber p osit iv e cases. Therefore, it is essential to deal with this data skew problem when training machine learning algorithms. This study uses two representative lung cancer datasets, PLCO an d NLST, wit h imb alan ce ratios (the proportion of samples in the majority class to those in the minority class) of 24.7 and 25.0, respectively, to predict lung cancer incidence. This research uses the performance o f 23 clas s imb alan ce methods (resampling and hybrid systems) with three classical classifiers (logistic regression, random forest, and LinearSVC) to identify the best imbalance techniques suitable for medical datasets. Resampling includes ten under-sampling methods (RUS, Etc.), seven over-sampling methods (SMOTE, Etc.), an d t wo integrated sampling methods (SMOTEENN, SMOTE-Tomek). Hybrid systems include (Balanced Bagging, Etc.). The results show that class imbalance learning can improve the classification abilit y o f t h e mo d el. Compared with other imbalanced techniques, under-sampling techniques have the highest standard deviation (SD), and over-sampling techniques have the lowest SD. Over-sampling is a stable met h od, an d the AUC in the model is generally higher than in other ways. Using ROS, the random forest p erforms t h e best predictive ability and is more suitable for the lung cancer datasets used in this study.

show abstract

Section: ) Data-level Methodsmentioning

confidence: 99%

A Comparative Performance Analysis of Data Resampling Methods on Imbalance Medical Data

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Fu et al (2016) presented an undersampling method based on Principal Component Analysis (PCA) and weighted comprehensive evaluation to improve the dataset's unbalanced condition during the forecast of software fault data. To overcome the problem with the under-sampling method, Jindaluang et al (2014) proposed a cluster-based under-sampling method which uses a clustering algorithm. This algorithm clusters the data in the majority class and selects a number of representative data in many proportions and then combines them with all the data in the minority class as a training set.…”

Section: Data Level Methodsmentioning

confidence: 99%

A Wide Scale Classification of Class Imbalance Problem and its Solutions: A Systematic Literature Review

Rekha¹,

Tyagi²,

Reddy³

2019

Journal of Computer Science

View full text Add to dashboard Cite

In today's world, most of the data (real world) is present in imbalanced form by nature. This is because of not having efficient algorithms to put this data (i.e., generated data by billion of internetconnected devices (IoTs)) in respective format. Imbalanced data poses a great challenge to (both) data mining and machine learning algorithms. The imbalanced dataset consists of a majority class and a minority class, where the majority class takes the lead over the minority class. Generally, several standard learning algorithms assume the balanced class distribution or equal misclassification costs. If prediction is performed by these learning algorithms on imbalanced data, the accuracy will be high for majority classes, i.e., resulting in poor performance. To overcome this problem (or improving accuracy of deision/prediction-making process), data mining and machine learning researchers have addressed the problem of imbalanced data using datalevel, algorithmic level and ensemble or hybrid methods. This article presents a systematic literature review and analyze the results of more than 400 research papers published between 2002-2017 (till June 2017), resulting in a broader and elaborate investigation of the literature in this area of research. Note that extension of this article/work will contain till December 2018 research articles, which will be published in June 2019 (now these more papers/articles did not include due to no. of pages/space issues). The systematic analysis of the research literature has focus on the key role of Data Intrinsic Problems in classification, handling the imbalanced data and the techniques used to overcome the skewed distribution. Furthermore, this article reveals patterns, trends and gaps in the existing literature and discusses briefly the next generation research directions in this area.

show abstract

“…al. [9] have discussed under-sampling using cluster based approach to balance data. Majority class samples are clustered and prominent samples are selected as the majority samples.…”

Section: A Data Level Approachmentioning

confidence: 99%

Rare Class Problem in Data Mining: Review

Dongre

2017

ijarcs

View full text Add to dashboard Cite

Class imbalance problem is getting so much attention of researchers now a days. In real life there are number of applications that generates imbalanced data sets. Imbalance nature of data makes classification task difficult. Dealing with these kinds of imbalanced dataset is the one of the biggest challenge in the data mining. Imbalanced dataset means the ratio of positive and negative classes is not balanced. The class that is having more number of samples is known as majority class and the class that is having less number of samples is known as minority class samples. Minority class samples are less but important. In the classification task, most of the times, we are ignoring minority class samples and more concentrating on majority class samples. This leads to good overall accuracy but poor minority class detection rate. Many algorithms have been proposed to deal with the imbalanced data problem but each has its prons and corns. Different techniques used for handling imbalance data are discussed here.

show abstract

Under-sampling by algorithm with performance guaranteed for class-imbalance problem

Cited by 12 publications

References 7 publications

A Comparative Performance Analysis of Data Resampling Methods on Imbalance Medical Data

A Comparative Performance Analysis of Data Resampling Methods on Imbalance Medical Data

A Wide Scale Classification of Class Imbalance Problem and its Solutions: A Systematic Literature Review

Rare Class Problem in Data Mining: Review

Contact Info

Product

Resources

About