Malicious JavaScript Detection via Browser Emulator Behavioral Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods are ineffective in detecting and blocking malicious JavaScript due to its obfuscation, leading to high false negatives and positives, and fail to identify websites that attempt to download malicious scripts, as they cannot distinguish between legitimate and malicious obfuscation.
Innovation Solution
A system that uses an enhanced browser emulator with instrumented functions to analyze and score website behaviors, flagging suspicious activities such as cookie theft, dynamic target generation, and shellcode insertion, and categorizing vulnerabilities to quantify the likelihood of hostile intent, employing dynamic analysis and behavioral tracing within a virtual machine environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If static analysis such as signature matching is used to detect malicious JavaScript, then detection speed is improved, but detection accuracy deteriorates due to obfuscation
Solution Approach 1:
The patent replaces static signature matching (mechanical pattern recognition) with dynamic behavioral analysis using a browser emulator. The system executes JavaScript in a controlled virtual environment and monitors actual runtime behaviors, substituting the mechanical search-for-signatures approach with a dynamic observation-of-execution approach that can distinguish obfuscated malicious code from legitimate code based on its actual actions rather than its appearance.
Solution Approach 2:
The patent changes the detection parameter from static code characteristics (signatures, patterns) to dynamic behavioral parameters (API calls, navigation actions, data modifications). By measuring what the code actually does during execution rather than what it looks like in source form, the system overcomes obfuscation while maintaining detection speed through automated behavioral scoring.
2Reliability
If obfuscation detection is used to identify malicious JavaScript, then potential threats are flagged, but false positives and false negatives increase
Solution Approach 1:
The patent performs preliminary execution of JavaScript in a controlled browser emulator environment before making security decisions. By pre-executing the code in a safe virtualized setting and observing its behavioral fingerprint, the system establishes a baseline of actual actions taken, which serves as more reliable evidence of malicious intent than mere presence of obfuscation techniques, thereby reducing false positives and negatives.
Solution Approach 2:
The system implements feedback through behavioral scoring, where observed actions during emulator execution are weighted and summed to produce an overall maliciousness score. This feedback mechanism allows the system to adjust its detection threshold and interpretation based on the cumulative evidence from multiple behavioral indicators, improving reliability while reducing error rates compared to binary obfuscation detection.
3Measurement precision
If behavioral analysis in a browser emulator is used to detect malicious JavaScript, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces a browser emulator as an intermediary layer between the suspicious JavaScript and the actual target system. This virtualized browser environment acts as a safe mediator that executes the potentially malicious code and captures its behavioral output, allowing accurate detection without requiring complex analysis of the JavaScript source itself or risking the target system. The intermediary handles the complexity of JavaScript execution internally while presenting simplified behavioral observations to the detection system.
4Measurement precision
If dynamic analysis with instrumented functions is used to trace JavaScript behavior, then vulnerability categorization is improved, but processing time increases
Solution Approach 1:
The patent applies partial monitoring by focusing instrumented function calls only on specific high-risk JavaScript APIs and behaviors most likely to indicate security vulnerabilities. Rather than tracing every single function call and operation, the system selectively instruments only the critical paths and dangerous function categories, achieving sufficient vulnerability categorization accuracy while limiting the processing time overhead to essential observations only.
Data Source
AI summary
An apparatus and system for scoring and grading websites and method of operation. An apparatus receives one or more Uniform Resource Identifiers (URI), requests and receives a resource such as a webpage, and observes the behaviors of an enhanced browser emulator as controlled by javascript provided by the webpage. The enhanced browser emulator tracks behaviors which when aggregated imply malicious intent.


