API Behavior Analytics Framework for Cross-Phase Threat Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional API security solutions fail to detect anomalies across the entirety of the attack chain, missing sophisticated API attacks that evolve over time and multiple phases, such as reconnaissance, resource development, and exfiltration, due to their focus on specific phases and types of vulnerabilities.
Innovation Solution
A behavior analytics system that collects and analyzes API request and response data using a neural-network-based embedding technology to create behavior fingerprints, clustering normal and abnormal user behavior, and employing specialized anomaly detection models to correlate attacks across different phases, providing a comprehensive view of API security threats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional solutions focus on detecting anomalies during specific phases and types of API security vulnerabilities, then detection precision for specific vulnerabilities is improved, but the ability to detect anomalies in the entirety of the attack chain deteriorates
Solution Approach 1:
The patent implements a universal behavior analytics framework that monitors and analyzes API calls across all phases of the attack chain (reconnaissance, resource development, initial access, execution, persistence, privilege escalation, defense evasion, credential access, lateral movement, command and control, and exfiltration). The system uses a unified anomaly detection engine that applies consistent behavioral analysis rules throughout the entire attack chain, enabling it to detect various types of API security vulnerabilities (authentication vulnerabilities, authorization vulnerabilities, data access vulnerabilities, etc.) with a single system, thereby achieving both detection precision and broad adaptability
2Measurement precision
If conventional technologies analyze only specific phases of the attack chain, then analysis depth for that phase is improved, but the completeness of the attack picture deteriorates
Solution Approach 1:
The patent segments the attack chain into distinct phases (reconnaissance, resource development, initial access, execution, persistence, privilege escalation, defense evasion, credential access, lateral movement, command and control, and exfiltration), with each phase having specialized detection rules and analysis depth. The behavior analytics framework processes API calls at each phase with appropriate analytical rigor while maintaining visibility across the entire attack chain, ensuring both deep analysis of individual phases and complete picture of the overall attack
3Adaptability or versatility
If the system monitors all API calls across all phases, then detection coverage is improved, but system complexity increases
Solution Approach 1:
The behavior analytics framework implements self-service mechanisms through automated baseline establishment, where the system automatically learns normal API call patterns from historical data and establishes behavioral baselines without manual intervention. The anomaly detection engine automatically adapts to new attack patterns and adjusts detection thresholds, reducing the need for complex manual configuration and maintenance while maintaining comprehensive detection coverage across all attack chain phases
Data Source
AI summary
A behavior analytics method for detecting and preventing different types of API based threats and attacks is disclosed. The method includes collecting request and response data of API calls from plurality of user sessions and storing it in a data lake. The method further includes extracting and combining features of the collected request and response data. The features may be associated with login behavior, API request content and behavior, API object accessing content and behavior, and API response content and behavior. The method also includes encoding the combined features via a neural network model to create a behavior fingerprint of each of the user sessions. Also, the method includes clustering the created behavior fingerprint to detect normal or abnormal user behavior. Thereafter, the method includes reporting the detected abnormal user behavior.


