A Fraud Detection Method for Low-Frequency Transaction

Zhao-hui, Zhang; Chen, Ligong; Liu, Min; Wang, Pengwei

doi:10.1109/access.2020.2970614

Cited by 21 publications

(8 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to Maniraj et al [14] "Credit Card Fraud Detection using Machine Learning and Data Science", is firstly passed the dataset through a local outlier factor and then through an isolation forest algorithm [15][16][17]. Then the authors applied the technique on the dataset obtained from a German bank in 2006.…”

Section: Literature Survey and Methodologiesmentioning

confidence: 99%

Online Transaction Fraud Detection Using Efficient Dimensionality Reduction and Machine Learning Techniques

Chenoori¹,

Kavuri²

2022

RIA

View full text Add to dashboard Cite

In recent years, there has been a rapid increase in the number of online transactions. Substantial growth has been reported in e-commerce and e-governance in the past few years. Due to this the number of people using online payment methods has also increased. This has led to an exponential rise in the number of transactions that happen every day. This increase in online transactions has further led to an increase in the number of frauds in the transactions. There is an ever-growing need to detect these fraudulent transactions as early as possible so that appropriate actions could be taken and losses due to these frauds could be minimized. This work proposes machine learning models which could use the previously known data and try to predict frauds based on information learned through the old data. We propose a statistical based dimensionality reduction technique and various machine learning models were tried for classification purpose. We experimented our proposed method on IEEE-CIS Fraud Detection dataset and the best results were obtained on the XGBoost model which is demonstrated in this paper.

show abstract

Section: Literature Survey and Methodologiesmentioning

confidence: 99%

Online Transaction Fraud Detection Using Efficient Dimensionality Reduction and Machine Learning Techniques

Chenoori¹,

Kavuri²

2022

RIA

View full text Add to dashboard Cite

show abstract

“…In [4] the problem with this approach is that random forest always leads to an over tting problem. [14][15][16][17][18][19][20][21][22][23][24][25] are about the implementation of the best performing models in the real-time scenario. According to these papers most of the real-time implementation is in the beginning stage.…”

Section: Related Workmentioning

confidence: 99%

Predicting Online Fraudulent Transaction Using Machine Learning

T.P¹,

Saravanan

et al. 2023

Preprint

View full text Add to dashboard Cite

Fraudulent online transactions have caused significant damage and loss to individuals and companies over a period of time. There has been an increase in online frauds with the progression of state-of-the-art technologies and worldwide communication. The design of efficient fraud detection algorithms is critical for reducing these losses. Machine learning and statistical techniques play a vital role in the detection of fraudulent transactions. Fraud detection model implementation is particularly challenging due to the lack of data, sensitive nature of data and the unbalanced class distributions. It is tough to draw inference and build better models due to the confidentiality of the records. In this paper we aim to focus on the particular problems: i) role of sampling in the presence of class imbalance (i.e. non fraudulent transactions are more in percentage of the total transactions), ii) to build and analyze various machine learning models, iii) to assess and validate the performances of different fraud detection techniques. This paper has research directions towards applying machine learning for data analysis. We have designed and assessed a prototype of a fraudulent transactions detection system which will be able to meet real-world demands and increase the security of transactions for the customers.

show abstract

“…The collection of such a dataset is done with the help of human loss prevention specialists who manually annotate data. Then, as the literature suggests, experts define numerical features using transaction data which are the base for classifier construction [43], [45]. Some attempts utilize only transaction frequency data [22].…”

Section: Leveraging Transactions Data For Incident Detection -Problem...mentioning

confidence: 99%

Variable-Length Multivariate Time Series Classification Using ROCKET: A Case Study of Incident Detection

Bier

Jastrzębska

Olszewski³

2022

IEEE Access

View full text Add to dashboard Cite

Multivariate time series classification is a machine learning problem that can be applied to automate a wide range of real-world data analysis tasks. ROCKET proved to be an outstanding algorithm capable to classify time series accurately and quickly. The textbook variant of the multivariate time series classification problem assumes that time series to be classified are all of the same length, while in realworld applications this assumption not necessarily holds. The literature of this domain does not pay enough attention to data processing pipelines for variable-length time series. Thus, in this paper, we present a thorough analysis of three preprocessing pipelines that handle variable-length time series that need to be classified with a method that requires the data to be of equal length. These three methods are truncation, padding, and forecasting of missing value. Experiments conducted on benchmark datasets, showed that the recommended procedure involves padding. Forecasting ensures similar classification accuracy, but comes at a much higher computational cost. Truncation is not a viable option. Furthermore, in the paper, we present a novel domain of application of multivariate time series classification algorithms, that is incident detection in cash transactions. This area poses substantive challenges for automated model training procedures since the data is not only variable-length, but also heavily imbalanced. In the study, we list various incident types and present trained classifiers capable to aid human auditors in their daily work. INDEX TERMS classification, incident detection, multivariate time series, ROCKET, varying-length time series I. INTRODUCTIONT IME series classification has become a vital domain of machine learning. The multitude of exciting real-life applications drives the development of the field and inspires fruitful research that aim at delivering new approaches, improving the existing ones, and adapting them to new types of data. This paper focuses on the task of classifying multivariate time series of unequal lengths. That is, each multivariate series can have a different length.The study presented in this paper is related to our project that deals with incident detection in cash transaction data. The business context of the project is loss prevention in quick-service restaurant operations. In our context, an incident is when a cashier (server, employee) is under-ringing items, voiding items, or performing other intentional operations that cause money losses for the restaurant.Thanks to many years of successful cooperation with our clients in the quick-service restaurant industry, we have accumulated an adequate database of transactions, and we rely on human experts to conduct incident audits. The humanauditor-based loss prevention scheme employed at the moment is the cornerstone of the daily operations of our company. However, there is a pressing need to automate audit tasks.We want to build an automated classifier that detects operations in which a cashier intentionally mishandles t...

show abstract

A Fraud Detection Method for Low-Frequency Transaction

Cited by 21 publications

References 24 publications

Online Transaction Fraud Detection Using Efficient Dimensionality Reduction and Machine Learning Techniques

Online Transaction Fraud Detection Using Efficient Dimensionality Reduction and Machine Learning Techniques

Predicting Online Fraudulent Transaction Using Machine Learning

Variable-Length Multivariate Time Series Classification Using ROCKET: A Case Study of Incident Detection

Contact Info

Product

Resources

About