Biblio

Cyber defense can no longer be limited to intrusion detection methods. These systems require malicious activity to enter an internal network before an attack can be detected. Having advanced, predictive knowledge of future attacks allow a potential victim to heighten security and possibly prevent any malicious traffic from breaching the network. This paper investigates the use of Auto-Regressive Integrated Moving Average (ARIMA) models and Bayesian Networks (BN) to predict future cyber attack occurrences and intensities against two target entities. In addition to incident count forecasting, categorical and binary occurrence metrics are proposed to better represent volume forecasts to a victim. Different measurement periods are used in time series construction to better model the temporal patterns unique to each attack type and target configuration, seeing over 86% improvement over baseline forecasts. Using ground truth aggregated over different measurement periods as signals, a BN is trained and tested for each attack type and the obtained results provided further evidence to support the findings from ARIMA. This work highlights the complexity of cyber attack occurrences; each subset has unique characteristics and is influenced by a number of potential external factors.

Cyber attacks occur on a near daily basis and are becoming exponentially more common. While some research aims to detect the characteristics of an attack, little focus has been given to patterns of attacks in general. This paper aims to exploit temporal correlations between the number of attacks per day in order to predict future intensity of cyber incidents. Through analysis of attack data collected from Hackmageddon, correlation was found among reported attack volume in consecutive days. This paper presents a forecasting system that aims to predict the number of cyber attacks on a given day based only on a set of historical attack count data. Our system conducts ARIMA time series forecasting on all previously collected incidents to predict the expected number of attacks on a future date. Our tool is able to use only a subset of data relevant to a specific attack method. Prediction models are dynamically updated over time as new data is collected to improve accuracy. Our system outperforms naive forecasting methods by 14.1% when predicting attacks of any type, and up to 21.2% when forecasting attacks of a specific type. Our system also produces a model which more accurately predicts future cyber attack intensity behavior.