Maxset Worldwide Inc. — Department of Defense SBIR Phase I: AF182-005

Description

Automated and Efficient Extract Transformation and Labeling of Textual Data in a practical, rigid and structured way. 200 million characters per hour deterministic formal machine translation of raw and ambiguous documents. • Common input file types include; pdf, doc, ppt, txt, csv, xls, html, etc. • Structuring is 100% accurate (hash value provable) • Structuring and labeling is immutably retained (JSON & CSV) • Results can be fed directly into other utilization tools • Extraction and search results are >97% accurate • Commercial competitors’ extraction and search results are less than 70% accurate • Results retain content, context and meaning for user (not just location) • Document integrity can be instantly and successively confirmed and validated • Offered for use on stand-alone machines (laptop/PC), servers (remote and on-premises) and cloud-based. Our unique on desk machine Windows OS applications ( legacy versions of Windows as well) produces best in class speeds. We also claim the novelty of taking raw text at better than expected speeds and create database ready output for N0SQL, SQL HDFS and HDF5 files systems at speeds 10000x faster than a human plus prescriptive accuracy.