Christopher Norman

Title: Christopher Norman and Graph-Inference-Based Text Extraction

Introduction

Christopher Norman is an inventor associated with Fagersta, Sweden. His work includes a patent for using graph inference and machine learning to extract text from unstructured documents. He is associated with Sciome, LLC, and has 1 patent listed in the information provided.

Latest Patents

Norman’s patent, “Methods and Systems for Graph-Inference-Based Text Extraction from Unstructured Documents,” describes a method for converting glyphs on a document page into text. The method identifies each glyph and its position, then uses that positional information to create an adjacency graph in which each node represents a glyph.

A machine-learning model classifies the graph’s nodes and edges. The glyphs are then grouped according to those classifications to produce text output. This approach applies graph-based analysis to the challenge of extracting text from unstructured documents.

Career Highlights

The available information associates Norman with Sciome, LLC, and identifies his patent as a contribution to document-processing technology. The described method combines glyph identification, spatial relationships, graph construction, and machine-learning classification.

Collaborations

Norman’s listed coworkers include Brian Howard and Ruchir Shah. The information provided does not specify their roles in the patent or the development of the described method.

Conclusion

Christopher Norman’s listed invention uses graph inference and machine learning to organize glyphs and extract text from unstructured documents, highlighting his contribution to document-processing innovation.