Matthew Sevrens

Title: Matthew Sevrens: Innovator in Neural Networks for Transaction Data

Introduction: Matthew Sevrens is a notable inventor based in Campbell, CA (US). He has made significant contributions to the field of information extraction through his innovative work on neural networks. With a focus on enhancing transaction data processing, Sevrens has developed methods that leverage deep learning techniques.

Latest Patents: Sevrens holds 1 patent for his invention titled "Neural networks for information extraction from transaction data." This patent discloses methods, systems, and computer program products that implement character-level deep neural networks for extracting information. The system he developed utilizes character-level information from transaction records to classify transactions and tag individual sections by entity type. It employs multiple character-level models and can identify entities such as service provider names using recurrent neural networks (RNNs), including long short-term memory (LSTM) models.

Career Highlights: Matthew Sevrens is currently associated with Yodlee, Inc., where he applies his expertise in neural networks to improve transaction data analysis. His work has been instrumental in advancing the capabilities of information extraction systems.

Collaborations: Sevrens collaborates with talented individuals in his field, including his coworker Zixuan Pan. Their combined efforts contribute to the innovative projects at Yodlee, Inc.

Conclusion: Matthew Sevrens is a pioneering inventor whose work in neural networks has the potential to transform how transaction data is processed and analyzed. His contributions to the field are noteworthy and continue to influence advancements in information extraction technologies.