Program source code for EA Mining from API SI Integrations

This file contains the complete source code for the <b>“EA Mining from APIs”</b> – a research prototype that derives high-level architectural views, expressed in the Archimate notation, directly from <i>unstructured</i> API-gateway log streams. Developed as part of an ongoing scientific project at the University of Trás-os-Montes e Alto Douro (UTAD), the program:<b>Ingests raw logs</b> from popular gateways (e.g., Kong, Apigee, NGINX, AWS API Gateway) without requiring prior schema or preprocessing.<b>Parses, clean and normalises</b> request/response metadata, automatically detecting services, endpoints, consumers, and routing paths.<b>Applies heuristic clustering &amp; lightweight NLP</b> to infer functional components, communication channels, and dependency directions hidden within the traffic.<b>Maps the findings to Ontologial concepts</b> (Application Component, Interface, Flow, etc.), building an in-memory model with temporal and frequency annotations.<b>Exports</b> the reconstructed view as an <code>.archimate</code>/<code>.xml</code> file (compatible with Archi, Modelio, and other tooling) or as a GraphML/CSV dataset for further analysis.<b>Includes unit tests and example datasets</b> under <code>tests/</code> and <code>samples/</code> to support peer review and replication of the published results.<b>Unzip the file and read the README.md for instructions on how to run the program.</b>