Reviewer Guideline

The Iraqi Journal of Artificial Intelligence and Machine Learning (IJAIML) relies on a double-blind peer-review process to maintain scientific excellence, integrity, and objectivity. Reviewers play a critical role in ensuring that published research offers original, technically sound, and high-impact contributions to the global AI and machine learning community.

1. Core Ethical Responsibilities

Reviewers are expected to adhere to the following principles:

  • Confidentiality: Every manuscript received for review must be treated as a confidential document. Reviewers must not share, discuss, or use any unpublished data or ideas contained within the manuscript.

  • Double-Blind Integrity: To preserve anonymity, do not attempt to identify the authors. If the text or datasets contain obvious identifiers that compromise the blind process, please notify the Editor immediately.

  • Conflict of Interest: Decline a review request if you recognize a potential conflict of interest, such as a recent collaboration, competitive commercial interest, or close personal/professional relationship with the authors or their institutions.

  • Objectivity: Reviews must be conducted objectively. Personal or subjective criticism of the authors is strictly unacceptable. All evaluations must be supported by clear, constructive academic arguments.

2. Technical Evaluation Criteria

When evaluating an AI/ML manuscript, please systematically assess the following components:

A. Originality and Scientific Merit

  • Does the manuscript present novel algorithms, architectures, architectures, frameworks, or significant practical applications of AI/ML?

  • Is the problem statement clearly defined, and does the research bridge a genuine gap in current state-of-the-art literature?

B. Methodology and Technical Correctness

  • Is the proposed AI/ML model, framework, or algorithm technically sound and clearly mathematically formulated?

  • Are the data preprocessing techniques, training setups, and validation strategies (e.g., cross-validation, training/testing splits) rigorous and correct?

  • Is the code or dataset sufficiently documented to allow for reproducibility?

C. Experimental Evaluation and Discussion

  • Are the evaluation metrics (e.g., accuracy, precision, recall, F1-score, execution time, computational complexity) comprehensive and appropriate for the specific task?

  • Are the baseline models used for comparative performance analysis relevant and up-to-date?

  • Do the experimental results explicitly support the authors' claims and conclusions?

D. Presentation and Readability

  • Is the paper written in clear, grammatically sound academic English?

  • Are all figures, flowcharts, block diagrams, and tables clear, relevant, and correctly captioned?

  • Is the manuscript correctly structured according to the journal's guidelines and referenced using the correct citation style?

3. Constructing the Review Report

Your formal feedback submitted via the OJS portal should be structured into two sections:

  1. Comments for the Authors: Provide constructive, detailed feedback. Highlight specific strengths first, followed by numbered points outlining mandatory major revisions, minor revisions, and suggestions for improvement. Avoid harsh or dismissive language.

  2. Confidential Comments to the Editor: Use this section to state any serious underlying concerns regarding plagiarism, duplicate publication, data fabrication, or ethical issues that should not be shared directly with the authors.

4. Final Recommendation Options

Reviewers must select one of the following official recommendations to assist the Editorial Board:

  • Accept As Is: The paper is technically flawless and requires no modifications.

  • Minor Revision: The manuscript has strong technical merit but requires minor textual clarifications, additional citations, or slight adjustments to figures/tables.

  • Major Revision: The paper shows promise, but requires substantial work before reconsidering—such as expanding the experimental evaluation, adding critical baseline comparisons, or significantly restructuring the technical methodology. (This usually triggers a second round of review).

  • Reject: The work lacks fundamental novelty, contains severe mathematical or methodological flaws, or falls entirely outside the journal’s scope.

5. Timeliness

Reviewers are given a specific timeframe (typically 2 to 3 weeks) to complete their evaluation. If an unexpected delay arises or you realize you cannot complete the review within the deadline, please inform the editorial office immediately so the manuscript can be reassigned without delaying the publication workflow.