Study on Class Imbalance Problem with Modified KNN for Classification

Sasirekha, R.; Kanisha, B.; Kaliraj, S.

doi:10.1007/978-981-16-7610-9_15

Cited by 3 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When a class tend to classify, it should be distributed equally to make prediction accurately. But when there is an occurrence of an unequal distribution of dataset is known as imbalanced data [2]. which can lead a model to give wrong predictions.…”

Section: B Study On Imbalanced Problemmentioning

confidence: 99%

Randomized Ensemble Approach with ID3 Algorithm For the Prediction datasets with Imbalance Problem

Sasirekha

Kanisha

2023

IJRITCC

View full text Add to dashboard Cite

Nowadays, it is significant to make accurate prediction model by handling imbalance problem. When the larger dataset has been used in the prediction model, that data should be classified into classes which gives ‘0’ and ‘1’ to indicate negative and positive results. While classifying this target value, the larger number of instances can reside in one class and the remaining lower number of instances can be stored in another class. Because of this unequal distribution of data, the machine can be biased and there is high possibility to give wrong predictions. An inaccurate Dataset leads to misprediction. Hence, the imbalanced prediction dataset has been taken. This paper gives a proper information on Randomized ensemble approach with ID3 classifier for the imbalanced prediction dataset.

show abstract

Section: B Study On Imbalanced Problemmentioning

confidence: 99%

Randomized Ensemble Approach with ID3 Algorithm For the Prediction datasets with Imbalance Problem

Sasirekha

Kanisha

2023

IJRITCC

View full text Add to dashboard Cite

show abstract

“…Specifically, fraud skewness which represents the majority fraud class over the non-fraud class has been a major concern to studies, as it affects the performance of fraud detection model. The Skewed fraud instances can have a bad influence on machine learning algorithms such as distance-based algorithms [8]. Previous efforts in tackling fraud involve developing rule-based expert systems, statistical methods, machine learning, and riskbased methods [9], [10].…”

Section: Introductionmentioning

confidence: 99%

Financial Fraud Detection Using Value-at-Risk With Machine Learning in Skewed Data

Usman,

Abdullahi,

Liping

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The significant losses that banks and other financial organizations suffered due to new bank account (NBA) fraud are alarming as the number of online banking service users increases. The inherent skewness and rarity of NBA fraud instances have been a major challenge to the machine learning (ML) models and happen when non-fraud instances outweigh the fraud instances, which leads the ML models to overlook and erroneously consider fraud as non-fraud instances. Such errors can erode the confidence and trust of customers. Existing studies consider fraud patterns instead of potential losses of NBA fraud risk features while addressing the skewness of fraud datasets. The detection of NBA fraud is proposed in this research within the context of value-at-risk as a risk measure that considers fraud instances as a worst-case scenario. Value-at-risk uses historical simulation to estimate potential losses of risk features and model them as a skewed tail distribution. The risk-return features obtained from value-at-risk were classified using ML on the bank account fraud (BAF) Dataset. The value-at-risk handles the fraud skewness using an adjustable threshold probability range to attach weight to the skewed NBA fraud instances. A novel detection rate (DT) metric that considers risk fraud features was used to measure the performance of the fraud detection model. An improved fraud detection model is achieved using a K-nearest neighbor with a true positive (TP) rate of 0.95 and a DT rate of 0.9406. Under an acceptable loss tolerance in the banking sector, value-at-risk presents an intelligent approach for establishing data-driven criteria for fraud risk management.

show abstract