Taking Advantage Of Machine Learning, Kaspersky Successfully Increases APT Detection By 25 Percent

Improved Persistent Threat Illustration (photo: Kaspersky)

JAKARTA - Kaspersky announced its success in increasing advanced persistent threat detection (APT) by 25 percent during the first half of 2024.

By utilizing machine learning techniques in its internal services, Kaspersky's Global Research and Analysis Team (GREAT) revealed that the thousands of new advanced threats target the government, finance, company and telecommunications sectors.

Machine learning models used in Kaspersky solutions use techniques such as Random Forest and term frequency inverse document frequency (TF-IDF) to process large amounts of data, which allows for faster and more accurate detection of subtle threats.

The combination of this ML method allows the identification of compromise indicators (IoCs) that traditional detection systems might ignore, leading to more precise anomaly detection and significant improvement in the overall threat detection capability.

The continued utilization of Kaspersky machine learning has allowed its system to process millions of data points every day, providing real-time insight into emerging threats.

Seeing this increase, Amin Hasbini as Head of the META Research Center at the Kaspersky GREAT was also surprised, because the results had exceeded their expectations.

"This technology increases the accuracy of detection and encourages proactive defense strategies, helping organizations to remain superior in dealing with the growing cyber threat," Amin said.

The fuller research results will be discussed at GITEX 2024, where Kaspersky will participate in a panel on the impact of AI on cybersecurity.