[Tool] Automatically Extracting Hardware Descriptions from PDF Technical Documentation

Abstract

The paper presents a modular processor designed to extract detailed data sets from technical documentation in PDF format, specifically targeting microcontroller hardware descriptions. The extracted data can be used for code generation in the embedded software domain. The table processing and text mining paradigms are applied to extract and convert data from the technical documentation in a deterministic process that yields reproducible results. The processor functionality is verified by comparing the data extracted from the technical documentation against the existing machine-readable sources. Most reviewers find the work very interesting and beneficial for the community.