Abstract
Mapping complex operational workflows across isolated, unstructured procedural articles is difficult for traditional process mining or graph extraction tools. Disclosed systems and methods implement a multi-stage deterministic pipeline to transform unstructured documents into a canonical, cycle-condensed directed acyclic graph (DAG). The system performs a single linear pass to extract a structured state contract formalizing procedural inputs and outputs for each document. An inverted index maps producer document outputs to consumer document inputs to synthesize the directed graph. The resulting graph and constituent nodes are processed to condense procedural cycles and evaluated via quantitative scoring models to identify operational bottlenecks, protect structural backbone procedures, and prioritize automation initiatives.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Rengarajan, Hariprasad; Agrawal, Bhasha; Kaul, Kunal; and Jindal, Rahul, "Synthesizing and Scoring Operational Process Graphs From Unstructured Documents", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/12368