Advanced security testing using a cyber‐attack forecasting model: A case study of financial institutions

Abstract

As the number of cyber‐attacks on financial institutions has increased over the past few years, an advanced system that is capable of predicting the target of an attack is essential. Such a system needs to be integrated into the existing detection systems of financial institutions as it provides them with proactive controls with which to halt an attack by predicting patterns. Advanced prediction systems also enhance the software design and security testing of new advanced cyber‐security measures by providing new testing scenarios supported by attack forecasting. This present study developed a model that forecasts future network‐based cyber‐attacks on financial institutions using a deep neural network. The dataset that was used to train and test the model consisted of some of the biggest cyber‐attacks on banking institutions over the past three years. This provided insight into new patterns that may end with a cyber‐crime. These new attacks were also evaluated to determine behavioral similarities with the nearest known attack or a combination of several existing attacks. The performance of the forecasting model was then evaluated in a real banking environment and provided a forecasting accuracy of 90.36%. As such, financial institutions can use the proposed forecasting model to improve their security testing measures.

A New Time Series Dataset for Cyber-Threat Correlation, Regression and Neural-Network-Based Forecasting

“…The highest value of MSE/MAPE was recorded for Australia on on-demand scan and the lowest value of the MSE/MAPE was identified for Saudi Arabia's local infection. Hence, the accuracy of predictions showcased with this considerably extended dataset markedly surpasses that of existing studies [22][23][24][25][26][27][28][29].…”

“…were performed, along with predictions related to cyber threats against randomly selected countries like Australia, New Zealand, Qatar, Saudi Arabia, and UAE. A thorough performance evaluation revealed a higher accuracy in predicting cyber threats, in comparison with existing studies [22][23][24][25][26][27][28][29].…”

“…In the last six years there have studies that have focused on the regression [17][18][19], correlation [20,21], and prediction [22][23][24][25][26][27][28][29] of cyber-attacks in order to boost cyber preparedness. However, because of the lack of a robust cyber-attack dataset, these existing studies faced limitations in terms of their prediction accuracy.…”

Technology-Oriented Innovations and Cyber Security Challenges in the Healthcare Delivery System: A perspective from Developing Economy

“…Nonetheless, all these theories focus on extrinsic rather than intrinsic factors which deal with matters like individuals' mindset of personal and professional lives explained by decision styles. So, many institutions especially, in developing countries are not adequately prepared to remedy the consequences of cyber-attack (Qasaimeh et al, 2022). However, to defend themselves against cyber criminals, institutions must be more prepared and conversant with cyber knowledge (Diogenes & Ozkaya, 2019).…”

“…Future research can adopt a different research method like mixed methodology or a wholly empirical method to yield more robust insights. Regarding the scope, firstly, the technology-driven innovations as described in this work stem from the works of (Aufegger et al, 2020;Banskota et al, 2020;Carney, 2011;Chitungo et al, 2021;Demirkan et al, 2020;Gleiss & Lewandowski, 2022;Qasaimeh et al, 2022). Safe to admit that, empirically validated data will be enough to affirm or prove otherwise, despite, the reluctance of some organizations to release data due to concerns for security perceived to be sensitive.…”