This inventor holds 1 USPTO granted patent. Top assignee: Fmr Corp.. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Mathew James Pazaris: Patent Innovation in Document Processing
Introduction
Mathew James Pazaris is an inventor based in Harvard, Massachusetts, whose work includes a patented method for representing the positions of textual objects in documents. His invention applies computer processing and machine learning to document images.
Latest Patents
Pazaris is associated with 1 patent, titled “Method for Adaptively Encoding Positions of Textual Objects in a Document.” The method receives an image of a document and determines the absolute position of a textual object within it. It then normalizes that position to obtain a relative position and calculates a bin size for dividing at least one image axis into evenly spaced bins. The relative position is discretized using those bins, and the resulting position, together with the object’s textual content, is provided as input to a machine learning model.
Career Highlights
Pazaris is associated with FMR Corp. His patent reflects an approach to encoding document layout information in a form that can be used by machine learning systems.
Collaborations
The supplied information identifies Chen Lin and Piush Kumar Singh as Pazaris’s coworkers. Their names are included here as professional associations; no specific joint patent work is provided.
Conclusion
Mathew James Pazaris’s patented method combines document-image analysis, adaptive position discretization, and machine learning input preparation, highlighting his contribution to innovation in document processing.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
