by
Dr. Shaista Anwar
Assistant Professor, Business department.
Research Overview
In the early years of the fast and active development of the 80s and 90s, financial frauds happened to be pretty simple. They were not more than duplicity of forged cheques, draining peoples or investors’ money through fake company formations, and minutest scrutinization of the loan documents, etc. These were the only means for the fraudsters to play around. It is only after the advancement and adoption of desktop culture we have witnessed the whole new age of cybercrime and digital frauds. Investors, retailers, businesses, and corporates none were spared and were hard hit. Since then, financial frauds have become intimidating for businesses and especially for banks across the globe.
With the advent and continuous advancement of technology, it has further complicated the ways and means for the fraudsters to end up with catastrophic consequences. As data is growing, many related connected challenges are also increasing. With the help of data science various aspect of data can be analyzed, various pattern of accessing the data can be understood, which can eventually help to understand the probability of risk associated with various pattern of storing / accessing / retrieving the data. This paper also presents an analysis on open source dataset, taken from Kaggel, for the data analysis by using logistic regression, and the results of which are measured with a confusion matrix, which provides a clearer understanding of the dataset.
This paper concluded that with the advent of technology the data is also growing continuously and to keep pace with work and life majority of work has been shifted to the online platforms. And during COVID-19 pandemic, more and more data and transactions have shifted to the digital world. At the same time, this is also a “dream come true” for hackers and scammers as more people are online. So this is a situation where more frauds can be executed, and that is where it is imperative to understand more about the patterns related to frauds. Different securing algorithms for data safety like watermarking and SVM algorithm can be used to make it more secure. Data science provides the flexibility to understand these kinds of patterns more specifically, which helps the reduction of these frauds.