Kevin Huang

Title: **Kevin Huang: Innovating Relation Extraction Technologies**

Introduction

Kevin Huang, an inventive mind based in Sunnyvale, CA, has made significant contributions to the field of relation extraction with his patented technologies. With a total of two patents under his name, Huang demonstrates a keen ability to blend computing advancements with practical applications that enhance language modeling and entity processing.

Latest Patents

Kevin Huang's most recent patents include “System and method for relation extraction with adaptive thresholding and localized context pooling” and “System for entity and evidence-guided relation prediction and method of using the same.”

The first patent outlines a computing device, complete with a processor and a storage device that holds computer-executable code. This system is designed to provide a document, embed entities within the document into embedding vectors, and predict the relationships between pairs of entities using an advanced language model. Notably, this relation prediction incorporates an adaptive threshold, determining the existence of a relation based on the logit function.

The second patent focuses on multitask prediction and describes similar computational elements, utilizing a head entity and document to process and predict head-tail relationships among entities. Through this innovative approach, the system efficiently extracts and combines relevant information to support evidence-driven relational predictions.

Career Highlights

Throughout his career, Kevin Huang has worked with notable companies such as Beijing Wodong Tianjun Information Technology Co., Ltd. and JD.com American Technologies Corporation. His experiences at these companies have significantly influenced his innovative output and contributed to his patented developments in relation extraction technologies.

Collaborations

In his professional journey, Kevin has collaborated with talented peers like Jing Huang and Guangtao Wang. These partnerships not only foster creativity but also enhance the quality and applicability of his inventions in the realm of language processing and entity recognition.

Conclusion

Kevin Huang is a remarkable inventor whose work in relation extraction showcases the fusion of technology and practical needs. With his patents addressing complex problems in language modeling, he continues to pave the way for future innovations in the field. His career accomplishments and collaborations only add to the impact he is making in the technological landscape.