International Research Journal of Engineering and Technology (IRJET) Volume: 11 Issue: 04 | Apr 2024
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
Machine Learning-Based Fraud Detection In Banking Transactions Siravuri Raghu Varma1, Kuppli Mokshit Srinivasa2, Nikhil3, K Adithya4 1,2,3,4 Student, GITAM(Deemed to be University), Visakhapatnam ,Andhra Pradesh, India. ---------------------------------------------------------------------***--------------------------------------------------------------------Abstract— Financial fraud detection is crucial across This section presents the net income of the company after
deducting expenses from revenues. An up-to-date picture of their assets, stakeholders and shareholders' stocks by the finance sheet. The cash flow statement evaluates how well a business generates enough cash to cover its debt payments, investments, and operating costs. Money notes are extra details that offer classification and more details regarding certain things.
sectors. This study presents a machine learning model to proactively identify fraudsters. It analyzes transactions to predict fraud, splitting data for precise training. Using machine learning, transactions are categorized as fraudulent or authentic. Results show high recall, accuracy, precision, and F1-score, indicating effective fraud prediction. This module offers a simple yet potent tool to curb financial fraud, saving costs and preserving integrity. We used algorithms like KNN algorithm, random forest algorithm, Adaboost algorithm and Decision tree classifier. Organizations can leverage such technology to detect and prevent fraudulent activities, safeguarding resources and reputation.
The openings of some events, assets, and modified account policies are among the topics covered in these notes. These disclosures are essential in order to support the money presented on the money statements. Bank transaction fraud is the act of manipulating money statements to make a company appear more bigger than it used to be, and boost stock values. In auditing, the money triangle serves as a route to illustrate the reasons behind a person's choice to commit fraud. The tri-components of the money triangle motivation and rationalization where all work together to promote fraud and raise money fraud.
Keywords— Transaction Analysis, KNN Algorithm, Random Forest, Ada Boost and Decision Tree Classifier.
Introduction 1.1 Introduction
This hypothesis has been widely applied by auditing experts to handle and to assess financial fraud, knowledge of the fraud triangle is essential. According to Gupta and Singh, the likelihood of fraud rises when there are incentives present, such as the need to meet goals or make up for losses. The business will face pressure or temptation to engage in dishonest business activities. In addition, the absence of inspections or ineffective controls creates a suitable environment for fraud. The process of rationalization occurs when the person aims to justify the fake action which could be affected by other people and their conditions.
The employment of dishonest, unlawful, or misleading practices to obtain financial benefits is known as financial fraud. Fraud can occur in a variety of financial contexts, such as banking, insurance, taxation, corporations, and more. A rising issue is fiscal fraud and evasion, which includes money laundering, tax evasion, cc fraud, and finance fraud. Even with efforts to eradicate financial fraud, many dollars are lost annually, which has a negative impact on society and industry. Banks, retailers, and individuals have all been severely impacted by this significant financial loss. These days, there is a marked increase in fraud efforts, which emphasizes the importance of fraud detection. Regarding the certified examiners, 10% of occurrences are involved.
The interactions and activities of a single, crucial system component—the performance analyst—are depicted in the image. They are the machine learning model's primary developers. Through data-driven insights, they play a crucial role in improving their performance, boosting efficiency, and accomplishing strategic goals. Their primary responsibilities include selecting data, loading it, preprocessing, separating it, classifying it, creating machine learning models that can anticipate and produce the needed results. The first step in the performance analyst/machine learning engineer's job is choosing and loading the dataset into the model. They can now read and comprehend the data.
Financial statement fraud is less common than asset misappropriation and corruption, yet the money consequences of these types of laundering cases are not that serious. When auditing, illustrating the reasons behind a person's choice to commit fraud. The three components of the money triangle, motivation, and rationalizing work together to promote fraudulent behavior and raise the risk. The study focuses more on money fraud. Money statements include information about a company's operations and money-wise performance like income rate earnings on the company.
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 118