Skip to main content

Evaluating the Performance of Two Hybrid Feature Selection Model of Machine Learning for Credit Card Fraud Detection on Classification and Prediction Methods | Chapter 06 | Advances in Applied Science and Technology Vol. 2


The hybrid and non-hybrid feature selection model aimed at predicting and classifying whether a transaction is fraudulent or non-fraudulent using machine learning approaches. The objective of this study was to use supervised learning framework to differentiate fraudulent and genuine transactions. The proposed hybrid model utilizes feature selection methods namely; the Principal Component Analysis (PCA) and PCA-Backward elimination with multiple linear regression and Reduced Error Pruning Tree classifier (RepT) using Python and WEKA. Through five experiments carried out in this study, the proposed approach has proven to be effective for eliminating redundant features in the dataset that does not have significant impact using PCA and Backward elimination method to optimize the predictive behaviour of the credit card transactions. Our first findings from the experimental results revealed that the RepT with PCA-Backward-Elimination prediction accuracy 87.37% is higher than that of multiple linear regressions with PCA 73.35% and PCA-Backward-Elimination 73.34%. Our second findings also revealed that the RepT with PCA-Backward-Elimination classification accuracy of 99.9368% is higher than that of multiple linear regressions with PCA 99.9122% and PCA-Backward-Elimination 99.9105%. The performance metrics measures on the classification model of the logistic regression with PCA-Backward-Elimination indicates that the hybrid model negates it with the expectation of maximization to minimization (99.9122% to 99.9105%) but returns the same results on the RepT decision tree classification in both cases of  99.9368%. The proposed hybrid feature selection method with machine learning algorithms outperforms the non-hybrid feature selection method with machine learning algorithms for classification and prediction accuracy.

Author(s) Details

Dr. Stephen Gbenga Fashoto
University of Swaziland, Kwaluseni, Swaziland.

Prof. Olumide Owolabi
University of Abuja, Abuja, Nigeria.

Mr Elliot Mbunge
University of Swaziland, Kwaluseni, Swaziland.

Dr. Andile Simphiwe Metfula
University of Swaziland, Kwaluseni, Swaziland.

Read full article: http://bp.bookpi.org/index.php/bpi/catalog/view/38/139/271-1
View Volume: https://doi.org/10.9734/bpi/aast/v2

Labels:

Comments

Post a Comment

Popular posts from this blog

Greening Regional Airports: A Vision for Carbon Neutral Infrastructure | Chapter 12 | Contemporary Perspective on Science, Technology and Research Vol. 3

 This study provides an overview of the energy demand of a regional airport, divided into individual time horizons. The electrification of aircraft systems raises the question of whether airports will be among the largest electricity consumers in our infrastructure in the future. Sustainability and especially emission reductions are significant challenges for airports that are currently being addressed. The Clean Sky 2 project GENESIS addresses the environmental sustainability of hybrid-electric 50-passenger aircraft systems in a life cycle perspective to support the development of a technology roadmap for the transition to sustainable and competitive electric aircraft systems. This article originates from the GENESIS research and describes various options for ground power supply at a regional airport. Potential solutions for airport infrastructure with a short (2030), medium (2040), and long (2050) time horizon are proposed. In addition to the environmental and conservation benefi...
Post a Comment
Read more

Alkali Element Modification of Glucose Molecules as a Method to Dissolve Cancer Cells | Chapter 12 | New Visions in Medicine and Medical Science Vol. 4

  The present study highlights about alkali element modification of glucose molecules as a method to dissolve cancer cells. The central regulation of the mechanisms governing cell proliferation has little effect on cancer cells. Cancer cells are entirely independent of the central command and divide and proliferate on their own, making it challenging to activate their response mechanism. Precisely, this is the reason why they are at risk to the health of humans and/or any biological entities. Instead of trying to reconnect the central command of the growth control mechanism to cancer cells that are already out of the range, we present a method of using the cancer cell’s own irresponsive and uncontrolled growth mechanism to their disadvantage and destroy the cancer cells. We found that this is achievable in an atomic/molecular level study of the glucose molecule, which is the primary food source used for growth and energy generation by all cells in the body, including the cancer cel...
Post a Comment
Read more

Diagnosis and Management of Scrapie in Post Mortem Goats by Immunohistochemistry | Chapter 6 | New Visions in Medicine and Medical Science Vol. 4

  The main objective of this study was to contribute to the annual passive surveillance for this pathology, since the goats are a species considered reservoir for the prion. A class of pathologies known as Transmissible Spongiform Illnesses (TSEs), which affect both humans and animals, includes scrapie, a neurological disease with a slow and deadly course that affects sheep, goats, and mouflons. Given that Immunohistochemistry (IHC) is the globally accepted technique for identifying classical scrapie in goats—a species thought to be a potential zoonotic disease reservoir. Chile in relation to Bovine Spongiform Encephalopathy (BSE), is classified as a disease-free country. Scrapie is actively monitored for the sheep, but passively for the caprine species, which is why this study was carried out in slaughtered goats in the central north of the country (IV Region), to provide information to the surveillance system for this disease. Fifty brains of goats older than 2 years were used, r...
Post a Comment
Read more