Archive Browsing VOLUME 8 ISSUE 12 JULY 2020

Credit Card Fraud Detection using Bayesian Optimized K-Nearest Neighbors
Abstract: Credit card fraud is one of the most important problems that financial institutions are currently facing. Although the technology has allowed to increase the security in the credit cards with the use of PIN keys, the introduction of chips in the cards, the use of additional keys such as tokens and improvements in the regulation of its use is also a necessity for banks, to act preventively against this crime. To act preventively, it is necessary to monitor in real time the operations that are carried out and have the ability to react in a timely manner against any doubtful operation that is performed. This paper presents an implementation of automatic credit card fraud detection system using Bayesian Optimized K-Nearest Neighbors on Kaggle dataset. The selection of proper attributes for reducing the training overhead and claiming higher accuracy for the fraud detection using soft computing. Performance evaluation is achieved using confusion matrix plot with accuracy, sensitivity and precision values.

Authors: Deepika Kanungo, Lokesh Parashar

File Name: Deepika_81200-20-102.pdf
 Download Abstract | Download Full Paper

Supervised Learning Methods for Predicting Diabetes: A Systematic Review of the Literature
Abstract: Artificial intelligence (AI) and its benefits in the field of medicine have generated a great revolution. For this reason, we want to identify the supervised learning methods (a sub-area of artificial intelligence) and the factors used for the prediction of diabetes that have been more significant in terms of technique (of which highlight decision tree and its derivatives) and results. For the identification of these methods, a systematic review of the literature was carried out. Machine learning methods were extracted from all the articles found to consider them as antecedents. There are several supervised learning methods that can predict diabetes in which some are hybrids and others pure, one better than others depending on the case study. Finally, after a review of the selected articles, the pre-processing stage in the development of these models is highlighted to achieve a higher precision score.

Authors: Manoj Niwariya, Dr. Anil Rajput, Dr. Shailesh Jaloree

File Name: Manoj_81200-20-101.pdf
 Download Abstract | Download Full Paper

© 2016 IJDACR Journal. All rights reserved.