David DeBarr

Title: Innovations of David DeBarr in Machine Learning and Document Metadata

Introduction

David DeBarr is an accomplished inventor based in Lynnwood, WA (US). He has made significant contributions to the field of machine learning, particularly in the area of document metadata prediction. His innovative approach leverages advanced algorithms to enhance the way documents are processed and understood.

Latest Patents

David DeBarr holds a patent for a system that triggers the execution of machine learning-based predictions of document metadata. This system predicts metadata attributes associated with documents using machine learning models. The document may represent an interaction between entities. The system trains machine learning models to predict scores indicating whether a token or a sequence of tokens in a document represents a metadata attribute. The metadata prediction is used to annotate the document and display it to users. The system receives user feedback via the user interface and uses this feedback to evaluate or retrain the model. Additionally, the system generates training data by receiving a set of annotated documents and comparing them against other documents to identify matching documents. It determines when to execute the machine learning-based metadata prediction based on steps of the document workflow executed by the system.

Career Highlights

David DeBarr is currently employed at DocuSign, Inc., where he continues to innovate and develop solutions that enhance document management and processing. His work has been instrumental in advancing the capabilities of machine learning applications in real-world scenarios.

Collaborations

David collaborates with Kaushik Narayanan, contributing to the development of cutting-edge technologies in the field of document processing and machine learning.

Conclusion

David DeBarr's work exemplifies the intersection of technology and innovation, particularly in the realm of machine learning and document metadata. His contributions are paving the way for more intelligent document management systems that can adapt and learn from user interactions.