Credit Card Fraud Detection using Local Outlier Factor and Isolation Forest

John, Hyder; Naaz, Sameena

doi:10.26438/ijcse/v7i4.10601064

Cited by 40 publications

(22 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As discussed in [42], the AUC metric for a credit card fraud transaction dataset in Table 2 are: (i) LOF 0.584; (ii) KNN 0.961 and i-Forest 0.951. In addition, in [40] there is a direct comparison between the LOF algorithm and the i-forest one using a financial dataset, which is different from the dataset used in our study. So, according to the authors of [40], the LOF method for outlier values features 0.28 precision and the i-forest method 0.02.…”

Section: Comparison Of Different Algorithms Datasets and Resultsmentioning

confidence: 99%

“…In addition, in [40] there is a direct comparison between the LOF algorithm and the i-forest one using a financial dataset, which is different from the dataset used in our study. So, according to the authors of [40], the LOF method for outlier values features 0.28 precision and the i-forest method 0.02. In order to be fair, a direct comparison of our results with the aforementioned studies as well as any other study it is neither just nor indicative because the dataset used in this study is not similar to the datasets used in other studies in terms of type or volume.…”

Section: Comparison Of Different Algorithms Datasets and Resultsmentioning

confidence: 99%

“…The i-forest method has been widely used for financial outlier detection. More specifically, the authors of [40] conducted an analysis for outlier detection using data originating from credit card transactions and compared the i-forest algorithm with LOF. In the same direction, and by also utilizing i-forest and LOF, the authors of [41] compare the efficiency of these methods for outlier detection in ATM transactions.…”

Section: Outlier Detection Algorithms and Related Workmentioning

confidence: 99%

See 2 more Smart Citations

An Advanced Abnormal Behavior Detection Engine Embedding Autoencoders for the Investigation of Financial Transactions

et al. 2021

View full text Add to dashboard Cite

Nowadays, (cyber)criminals demonstrate an ever-increasing resolve to exploit new technologies so as to achieve their unlawful purposes. Therefore, Law Enforcement Agencies (LEAs) should keep one step ahead by engaging tools and technology that address existing challenges and enhance policing and crime prevention practices. The framework presented in this paper combines algorithms and tools that are used to correlate different pieces of data leading to the discovery and recording of forensic evidence. The collected data are, then, combined to handle inconsistencies, whereas machine learning techniques are applied to detect trends and outliers. In this light, the authors of this paper present, in detail, an innovative Abnormal Behavior Detection Engine, which also encompasses a knowledge base visualization functionality focusing on financial transactions investigation.

show abstract

Section: Comparison Of Different Algorithms Datasets and Resultsmentioning

confidence: 99%

Section: Comparison Of Different Algorithms Datasets and Resultsmentioning

confidence: 99%

Section: Outlier Detection Algorithms and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

An Advanced Abnormal Behavior Detection Engine Embedding Autoencoders for the Investigation of Financial Transactions

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Thus, they stated that IForest achieved AUROC 70% in high dimensional data while their dataset does not contain anomalies in the training samples. Meaning that they tested the models under semi-supervised anomaly detection type contrary to the authors in [16], [18].…”

Section: A Anomaly Detectionmentioning

confidence: 98%

“…On the other hand, the authors in [16], [17] used Local Outlier Factor (LOF) and Isolation Forest (IForest) techniques to detect the anomaly in a large scale data. Moreover, they used F1-score, Precision, and Recall as performance metrics except for Galante in [17] who added One-Class Support Vector Machines (OCSVM) as an additional technique to detect anomalies and he added AUROC as performance metrics.…”

Section: A Anomaly Detectionmentioning

confidence: 99%

Assembly Line Anomaly Detection and Root Cause Analysis Using Machine Learning

Abdelrahman

Keikhosrokiani

2020

IEEE Access

View full text Add to dashboard Cite

Anomaly detection is becoming widely used in Manufacturing Industry to enhance product quality. At the same time, it plays a great role in several other domains due to the fact that anomaly may reveal rare but represent an important phenomenon. The objective of this paper is to detect anomalies and identify the possible variables that caused these anomalies on historical assembly data for two series of products. Multiple anomaly detection techniques were performed; HBOS, IForest, KNN, CBLOF, OCSVM, LOF, and ABOD. Moreover, we used AUROC and Rank Power as performance metrics, followed by Boosting ensemble learning method to ensure the best anomaly detectors robustness. The techniques that gave the highest performance are KNN, ABOD for both product series datasets with 0.95 and 0.99 AUROC respectively. Finally, we applied a statistical root cause analysis on the detected anomalies with the use of Pareto chart to visualize the frequency of the possible causes and its cumulative occurrence. The results showed that there are seven rejection causes for both product series, whereas the first three causes are responsible for 85% of the rejection rates. Besides, assembly machines engineers reported a significant reduction in the rejection rates in both assembly machines after tuning the specification limits of the rejection causes identified by this research results.

show abstract