A comparative study on sampling techniques for handling class imbalance in streaming data

Nguyen, Hien M.; Cooper, Eric W.; Kamei, Katsuhiko

doi:10.1109/scis-isis.2012.6505291

Cited by 32 publications

(14 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The following methods are employed for sampling process in this work: 1). Resample is a method develops a unique dataset, which is replaced with a sample [30].…”

Section: International Journal Of Recent Technology and Engineering (mentioning

confidence: 99%

Predicting Student Academic Performance with Ensemble Classification Method on Imbalanced Educational Data

Chowdary*,

Prasanna,

Krishna T

et al. 2019

IJRTE

View full text Add to dashboard Cite

Education benefits a person in sustaining his present and future by assuring the goal of life. At present universities and colleges are mainly focusing to improve the academic performance of the students. Recently, many studies have concentrated on employing several machine learning models in the field of higher education to assist both the teachers and the students to identify their problems and can take remedial measures to improve their performances. Some of the earlier studies have been discussed about class imbalance problem and achieved poor prediction outcomes due to low performance of the classifiers. In this paper, we aim to improve the classification/prediction outcomes with a rule-based ensemble model based on various sampling strategies by addressing the class imbalance problem. The dataset used for this study has been collected from Vignan’s Lara and Vignan’s Nirula institutions based on considering various factors such as Attendance Percentage, No of Backlogs, Adjustable Nature, Concentration, Result History, Discipline in class, Usage of Social Media, Degree of Intelligence, Understanding of Subjects, Event Participation, Time Management, Extra Classes, Alternative Learning Skills, Logical Thinking, Bad Habits, Parents Education, Health Condition, Planning for higher studies, Family Support, Time Management, and Aggregate. To evaluate the efficiency, we also considered and compared our original dataset with different benchmark datasets and the performance measures of the proposed method is also tested with various sampling methods based on a learning rate parameter ranging between 0.1 and 0.8. The original data set with the re-sampling method with the proposed method achieved maximum precision values at a learning rate 0.3 with an accuracy rate of 98.36%. Finally, the obtained results were also compared with several baseline classifiers like Naïve Bayes, SVM, MLP, KNN, and OneR on the collected original datasets.

show abstract

“…The following methods are employed for sampling process in this work: 1). Resample is a method develops a unique dataset, which is replaced with a sample [30].…”

Section: International Journal Of Recent Technology and Engineering (mentioning

confidence: 99%

Predicting Student Academic Performance with Ensemble Classification Method on Imbalanced Educational Data

Chowdary*,

Prasanna,

Krishna T

et al. 2019

IJRTE

View full text Add to dashboard Cite

show abstract

“…Oversampling 20,21 and undersampling 22 are often adopted to solve the problem of sample imbalance. Such methods usually copy the small samples till the number is same as that of large samples or reduce the number of large samples, which easily causes data redundancy or loss of partial information.…”

Section: Fine-tuning and Classification Based On Aos-softmaxmentioning

confidence: 99%

Mechanism of situation element acquisition based on deep auto-encoder network in wireless sensor networks

Zhu

Yue

Song

et al. 2017

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

In order to reduce the time complexity of situation element acquisition and to cope with the low detection accuracy of small class samples caused by imbalanced class distribution of attack samples in wireless sensor networks, a situation element extraction mechanism based on deep auto-encoder network is proposed. In this mechanism, the deep autoencoder network is introduced as basic classifier to identify data type. In hierarchical training of the auto-encoder, a training method based on cross-entropy loss function and back-propagation algorithm is proposed to overcome the problem of weights updating too slow by the traditional variance cost function, and the momentum factors are added to improve the convergence performance. Meanwhile, in the stage of fine-tuning and classification of the deep network, an active online sampling algorithm is proposed to select the sample online for updating the network weights, so as to eliminate redundancy of the total samples, balance the amounts of all sample types, and improve the classification accuracy of small sample. Through the simulation and analysis of the instance data, the scheme has a good accuracy of situation factors extraction.

show abstract

“…OU is also effective compared to (i) other undersampling and oversampling methods (Nguyen, Cooper, and Kamei 2012) and (ii) various types of bootstrap aggregation, boosting, and hybrid ensemble data rarity methods used in the data analytics literature (Galar, Fernández, Barrenechea, Bustince, and Herrera 2012). Nguyen et al (2012) suggests that OU improves performance not only because it makes the ratio of minority (fraud) to majority (non-fraud) observations more balanced, but also because it more efficiently incorporates potentially useful majority cases. By increasing the balance between fraud and non-fraud cases, OU adjusts the focus of the classification algorithms towards the fraud cases.…”

Section: Preprint Accepted Manuscript -7 -mentioning

confidence: 99%

Finding Needles in a Haystack: Using Data Analytics to Improve Fraud Prediction

Perols

Bowen

Zimmermann

et al. 2016

The Accounting Review

143

View full text Add to dashboard Cite

Developing models to detect financial statement fraud involves challenges related to (1) the rarity of fraud observations, (2) the relative abundance of explanatory variables identified in the prior literature, and (3) the broad underlying definition of fraud. Following the emerging data analytics literature, we introduce and systematically evaluate three data analytics preprocessing methods to address these challenges. Results from evaluating actual cases of financial statement fraud suggest that two of these methods improve fraud prediction performance by approximately 10 percent relative to the best current techniques. Improved fraud prediction can result in meaningful benefits, such as improving the ability of the SEC to detect fraudulent filings and improving audit firms' client portfolio decisions.

show abstract

A comparative study on sampling techniques for handling class imbalance in streaming data

Cited by 32 publications

References 16 publications

Predicting Student Academic Performance with Ensemble Classification Method on Imbalanced Educational Data

Predicting Student Academic Performance with Ensemble Classification Method on Imbalanced Educational Data

Mechanism of situation element acquisition based on deep auto-encoder network in wireless sensor networks

Finding Needles in a Haystack: Using Data Analytics to Improve Fraud Prediction

Contact Info

Product

Resources

About