Zhizhou Song

Title: Zhizhou Song and His Work in Malicious-Code Detection.

Introduction

Zhizhou Song is an inventor associated with Kingston, California. His patent work focuses on using machine learning to identify malicious code and detect code clones.

Latest Patents

Song is credited with one patent, titled “Method and Apparatus for Clone Search.” The invention describes a method and apparatus for identifying malicious code by comparing a new code segment with known malicious-code segments. It converts code segments into vectors and uses cosine similarity to identify clones. The patent also describes neural-network training techniques designed to handle very long code sequences and obfuscated malicious code.

Career Highlights

Song is associated with BlackBerry Corporation. His patent reflects work at the intersection of machine learning, code analysis, and cybersecurity.

Collaborations

Song’s listed coworkers include Honghui Ding and Yuan Tian.

Conclusion

Zhizhou Song’s patent work addresses the challenge of detecting malicious code by applying vector comparisons and neural-network techniques to code analysis.