Research Data and Code for Adaptive Closed-Loop Digital Evidence Acquisition
Description
This dataset contains the computational research data, experimental results, configuration information, analysis outputs, and research prototype code supporting the study “Adaptive Closed-Loop Digital Evidence Acquisition in Dynamic Distributed Evidence Environments: A Prediction-Error-Driven Control Framework.” The materials include results from the principal controlled experiments, integrated ablation experiments, controlled action-competition experiments, and dynamic evidence-topology experiments. The dataset supports examination of prediction-error feedback, acquisition-behaviour adaptation, acquisition-induced disturbance, dynamic topology updating, and subsequent acquisition decisions. The experiments were conducted in a controlled deterministic simulation environment using synthetic representations and do not contain real-world forensic case data, personally identifiable information, or live digital evidence. The materials are provided to support transparency, reproducibility, and independent examination of the reported computational results.
Files
Steps to reproduce
Download the dataset and research code from the published Mendeley Data record. Review the README.md file in the repository root for the repository structure, experimental record, interpretation of results, and reproduction notes. Review the files in the documentation directory, particularly the prototype specification, experimental protocol, validation reports, and P16/P16B/P17 analysis reports. Install the Python dependencies listed in code/requirements.txt in a compatible Python environment. Review the source files in the code/aeac/ directory and the associated execution scripts to understand the deterministic AEAC prototype and simulation workflow. Use the supplied experimental result files in the data/ directory to reproduce the reported analyses. The P16 dataset contains the principal controlled experiment records, while P16B contains the separately identified action-competition and dynamic-topology mechanism experiments. The P17 files contain the final aggregation, paired comparisons, hypothesis assessment, and summaries used to interpret the experimental results reported in the manuscript. The experiments use synthetic representations in a controlled deterministic simulation environment. The repository does not contain real-world forensic case data, personally identifiable information, or live digital evidence. The supplied MANIFEST.json provides SHA-256 checksums for the repository files to support integrity verification. The reported experimental record comprises 1,050 principal controlled runs, 30 controlled action-competition runs, and 30 dynamic-topology runs, for a total of 1,110 run records.
Institutions
- Aditya UniversityAndhra Pradesh, Sūrampālem