Academic Integrity Review Report: "Attribute-Based Data Sharing with Flexible and Direct Revocation in Cloud Computing"

Academic fraud report · Geng Detector

Summary

This AI-assisted review concludes the paper is CLEAN (no academic misconduct detected). The work is a theoretical cryptography paper on attribute-based encryption (ABE), not a wet-lab biomedical study, so most image-based manipulation checks are inapplicable. The figures consist only of a system architecture diagram (Fig. 1) and performance comparison line charts (Fig. 2–3), which do not contain microscopy or blot imagery susceptible to splicing. Tables 1–4 present symbolic/theoretical comparisons rather than raw numeric data with last digits, so last-digit statistical tests cannot be triggered. Hardware/software timeline is fully consistent: published 2014-11-30 (received 2014-05), uses PBC library 0.5.12 and an Intel Xeon E3-1230 V2 CPU (released Q2 2012), which matches the era. The paper is presented as a journal extension of a 2013 IEEE INCoS conference paper with clearly enumerated additions. Funding IDs (e.g., 61402366, 2013ZX03002004) conform to Chinese NSFC numbering conventions. Confidence is moderate-to-high for non-applicability of biomedical fraud checks, but limited by the absence of raw experimental datasets to verify.

Verdict

CLEAN — No academic misconduct indicators identified. The paper is a theoretical cryptography contribution; biomedical image-manipulation checks and raw-data last-digit tests are largely inapplicable, and within the applicable checks the paper shows no anomalies.

Key findings

No image manipulation scope: The only figures are a generic system architecture schematic (Fig. 1) and performance line charts (Fig. 2, Fig. 3). No Western blots, microscopy, flow cytometry, or other wet-lab imagery is present.
No raw numeric tables: Tables 1–4 contain symbolic/theoretical parameters and formula-based comparisons, not raw measurements with last digits; last-digit randomness tests cannot be triggered.
Hardware/software timeline is internally consistent: Published 2014-11-30 (received 2014-05); experiments use PBC library version 0.5.12 on an Intel Xeon E3-1230 V2 CPU (released Q2 2012), which is fully consistent with a 2014 experiment.
Legitimate journal extension: The paper is described as an extended version of a 2013 IEEE INCoS conference paper, with explicitly enumerated additions (system architecture, detailed proofs, experimental results) — consistent with normal conference-to-journal expansion practice.
Methodological consistency: Algorithmic descriptions and comparison dimensions in Tables 3–4 are supported by corresponding mathematical derivations in the body.
Funding identifiers plausible: Six NSFC and other grants are listed (e.g., 61402366, 2013ZX03002004), conforming to Chinese funding-number conventions and to the authors' institutional timelines.

Evidence highlights

DOI: 10.3837/tiis.2014.11.021
Journal: KSII Transactions on Internet and Information Systems, Vol. 8, No. 11, published 2014-11-30.
Section 6.2 ("Performance Comparison", p. 4046) reports the PBC 0.5.12 / Intel Xeon E3-1230 V2 environment; both are verifiable as era-appropriate.
Figures present: Fig. 1 (architecture diagram), Fig. 2–3 (performance line charts). No high-risk imagery detected.
Tables present: Tables 1–4 (symbolic/theoretical comparisons; no raw last-digit datasets).

Notes

Scope limitation: This tool's core detectors target biomedical image reuse, splicing, and last-digit fabrication. Their non-trigger here reflects the paper's theoretical nature rather than positive confirmation that no fabrication occurred; mathematical proofs were not independently re-derived.
Confidence: Moderate-to-high for the inapplicability verdict on biomedical checks; lower for cryptographic correctness, which is out of scope for this review.
Final determination of academic misconduct requires official institutional investigation; this AI-assisted report is for discussion only and may contain false negatives.