Integrity Assessment: Improving the classification performance of microplastics by noise reduction and baseline correction of Raman spectra with a neural network-based algorithm (Optics Express, DOI: 10.1364/OE.597337)

Academic fraud report · Geng Detector

Summary

Verdict: Questionable (yellow). The report identifies two issues in the paper by Jiajin Chen, Weixiang Huang, Ligang Shao et al. (Optics Express, Vol. 34, No. 9, DOI: 10.1364/OE.597337). First, a missing upper bound appears in Section 2.3 (Page 17242) where the range for the number of Raman peaks N is written as '[5, ]', likely a LaTeX/typesetting artifact or incomplete edit. Second, a global random seed of 19580920 is set in Section 2.4 (Page 17243), which corresponds to the founding date of the University of Science and Technology of China (USTC), 20 September 1958; this is interpreted as a harmless author 'easter egg' that indirectly supports the authenticity of the experimental code. No core data fabrication or logical contradictions were found. The verdict is limited to textual and editorial concerns and does not constitute evidence of academic misconduct.

Verdict

Questionable (yellow). The paper shows minor editorial/proofreading issues but no clear evidence of data fabrication, image manipulation, or substantive research misconduct. Concerns are confined to typographical completeness and an authorial easter egg in code.

Key findings

Incomplete numeric range (Section 2.3, Page 17242, below Equation 4): The description of parameter N is given as "randomly selected within the range of [5, ]." The upper bound is missing, leaving an empty slot inside the interval.
Random seed 'easter egg' (Section 2.4, Page 17243): Global random seed is fixed at 19580920, which matches USTC's founding date (1958-09-20). Interpreted as a low-risk author signature rather than an integrity issue.
No image manipulation, statistical anomalies, duplicated figures, or fabricated datasets were identified in this report.

Evidence highlights

Location 1 — Section 2.3, Simulation data generation, Page 17242: Equation 4 explanation reads "$N$ represents the total number of Raman peaks, which is randomly selected within the range of [5, ]." The interval [5, ] lacks an upper bound value; possible LaTeX compilation error or incomplete edit.
Location 2 — Section 2.4, Model training, Page 17243: Explicit statement "the global random seed was fixed at 19580920." Cross-reference: USTC (the authors' affiliated institution) was established on 20 September 1958, consistent with the encoded date.
Severity: Finding 1 marked yellow (editorial sloppiness); Finding 2 marked yellow as a positive signal supporting code authenticity.

Notes

The DOI 10.1364/OE.597337 and page reference (oe-34-9-17239.pdf, Page 17242–17243) are reproduced exactly as in the source report.
Limitations: This assessment is based on the textual indicators reported by the original 'Geng' analysis; it does not include independent re-inspection of spectra, code, or figures. A formal conclusion of misconduct would require institutional investigation and access to the underlying data and code.
Recommended action is limited to contacting the authors for clarification regarding the missing upper bound in the [5, ] interval; no journal or institutional complaint is indicated based on current evidence.
Disclaimer: This summary is AI-assisted and intended for academic discussion only.