Class Imbalanced Data: Open Issues and Future Research Directions

Rekha, Gillala; Tyagi, Amit Kumar; Sreenath, N.; Mishra, Shashvi

doi:10.1109/iccci50826.2021.9402272

Cited by 13 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The handling of imbalanced data can be categorized into three groups: data-level methods, algorithm-level methods, and hybrid approaches (Santhi & Reddy, 2019;Niaz et al, 2022;Rekha et al, 2021;Johnson & Khoshgoftaar, 2019). Data-level methods involve modifying the dataset's samples to balance the distribution, achieved through oversampling (generating new samples for the minority class) and undersampling (removing samples from the majority class).…”

Section: Handling Of Imbalanced Datamentioning

confidence: 99%

Machine Learning Approach in Predicting Fraudulent Job Advertisement

Mohd Hanif,

Maarop,

Kamaruddin

et al. 2024

IJARBSS

View full text Add to dashboard Cite

As the world population grows, the demand for workers increases, leading to a rise in online job advertisements to connect employers with potential employees on a national scale. However, this shift also brings the risk of falling victim to fraud. Reported commercial crimes in Malaysia saw a 15.3% increase in 2021, with fraud being the highest among them. Several studies have proposed Machine Learning models to classify genuine and fraudulent job advertisements, but the analysis of certain techniques remains limited. The paper aims to develop a predictive model for identifying fraudulent job advertisements using selected features from imbalanced and balanced datasets. The Employment Scam Aegean Dataset was utilized to build Machine Learning classification models using Logistic Regression, Support Vector Machine, Decision Tree, and Naïve Bayes algorithms. These models were combined with different vectorizers like Term Frequency-Inverse Document Frequency, Bag of Words, and Hash. The Decision Tree model with Bag of Words vectorizer on a balanced dataset outperformed other models, achieving an accuracy of 0.705, precision of 0.73, recall of 0.70, F1-score of 0.71, and Area Under Curve score of 0.68. This model shows promise in effectively identifying fraudulent job advertisements, safeguarding job seekers from scams in the online job market.

show abstract

Section: Handling Of Imbalanced Datamentioning

confidence: 99%

Machine Learning Approach in Predicting Fraudulent Job Advertisement

Mohd Hanif,

Maarop,

Kamaruddin

et al. 2024

IJARBSS

View full text Add to dashboard Cite

show abstract

“…Class imbalance is a common concern associated with many real-world datasets [7] [14], and is usually exhibited by the skewness in frequency distribution towards the dominating label(s). Figure 1 highlights that labels such as cs.AR occur relatively infrequently than others (e.g: cs.AI).…”

Section: Addressing Class Imbalancementioning

confidence: 99%

An Empirical Study of Supervised and Active Learning Methodologies for Classifying Research Papers

Singh¹,

Dhani²,

Patel³

et al. 2023

Preprint

View full text Add to dashboard Cite

Document classification is one of the predominant tasks in Machine Learning, that by and large involves the use of supervised methodologies to determine most suitable categories for a given set of documents. In this project, we seek to conduct an empirical study, that, in addition to the conventional machine learning methods, experiments with active learning and semi-supervised methods for assorting a mix of labelled and unlabelled computer science scholarly articles hosted on the arXiV repository. We develop a solution using the gathered labelled data points, by experimenting with various multi-class supervised learning models such as K Nearest Neighbours, Naive Bayes, Random Forest, and Logistic Regression. We also account for the issue of class imbalance, and thus conduct experiments to determine a suitable oversampling methodology for addressing this concern. Lastly, we extend beyond the conventional machine learning classification methods to incorporate the concept of active learning, a type of iterative supervised learning, used when the unlabelled data is in abundance.

show abstract

“…In recent years, imbalanced data learning has become a significant challenge in the field of data mining [18]. Many imbalanced data learning methods have been proposed to address this challenge, such as most commonly SMOTE and its variants.…”

Section: Related Workmentioning

confidence: 99%

An improved SMOTE based on center offset factor and synthesis strategy for imbalanced data classification

Zhang

Deng

Huang

et al. 2023

Preprint

View full text Add to dashboard Cite

It is an enormous challenge for imbalanced data learning in the field of machine learning. To construct balanced datasets, oversampling techniques have been studied extensively. However, many oversampling methods suffer from introducing noisy samples and blurring classification boundaries, which leading to overfitting. To solve this problem, this paper proposes a new oversampling method, namely CS-SMOTE, for synthesizing minority class samples by three-point interpolation. CS-SMOTE is mainly based on the center offset factor and a synthesis strategy. First, the CS-SMOTE method removes noise samples, calculates the center offset factor, and selects sparsely distributed minority class samples by using the K-distance graph technique. Next, new samples are generated basing on sparse minority samples, random minority samples, and sub-cluster centers located in the same sub-cluster samples. Finally, multiple comparative experiments on 18 well-known datasets demonstrate the effectiveness and general applicability of the proposed CS-SMOTE method for the imbalanced data classification. The experiments show that CS-SMOTE outperforms other competitors in terms of classification accuracy, while avoiding the issue of overfitting.

show abstract

Class Imbalanced Data: Open Issues and Future Research Directions

Cited by 13 publications

References 31 publications

Machine Learning Approach in Predicting Fraudulent Job Advertisement

Machine Learning Approach in Predicting Fraudulent Job Advertisement

An Empirical Study of Supervised and Active Learning Methodologies for Classifying Research Papers

An improved SMOTE based on center offset factor and synthesis strategy for imbalanced data classification

Contact Info

Product

Resources

About