1. Submission and Production Guidelines

To support an uncompromised, objective, and timely evaluation, authors must comply with the following technical and operational submission protocols:

• Submission Routing: Manuscripts must be submitted directly via email to the Editor-in-Chief at akhand@cse.kuet.ac.bd while the automated online submission portal remains under development. Submissions must align precisely with the journal’s focus on applied data science, artificial intelligence, machine learning, optimization, and computational intelligence.
• Format and Layout: Manuscripts must be submitted in Microsoft Word format (.docx). Authors must use standard, universally accessible typefaces (such as Times New Roman) to prevent typographical artifacts or parsing distortions across diverse operating systems and review environments.
• Referencing Conventions: All citations, bibliographies, and reference arrays must strictly follow the APA referencing format.
• Anonymity & Dual-File Submission: To protect the integrity of the double-blind peer review process, the submission must be provided as two distinct, decoupled files:
  • Covering Letter File: Details the manuscript title, manuscript category (original research, review, data-resource paper, case study, or methodological contribution), a statement detailing the novelty of the research, comprehensive funding disclosures, acknowledgments, and an academic biography/CV of each contributing author.
  • Anonymized Manuscript File: Contains the complete technical narrative—including title, abstract, keywords, figures, tables, algorithmic pseudocode, mathematical derivations, empirical evaluations, and reference lists—completely stripped of author names, affiliations, email addresses, institutional identifiers, or document metadata.

2. Code of Ethics for Authors

Originality, Plagiarism, and Text Recycling

JADSCI considers only entirely original scientific and methodological contributions. Duplicative and derivative publication models are strictly forbidden:

• Plagiarism & Redundant Publication: Reformulating previous benchmarks, rotating identical dataset evaluations across slight model variations, or repackaging published findings to generate artificial paper yields degrades academic trust. Submissions found to duplicate previous works will be rejected summarily.
• Text Recycling (Self-Plagiarism): Authors are prohibited from repurposing substantive narrative sections, experimental setups, or theoretical derivations from their own prior publications without explicit citation and appropriate attribution.
• Single Submission Guarantee: Authors must explicitly affirm in their submission correspondence that the manuscript has not appeared previously in any peer-reviewed venue (journal, conference proceeding, or book chapter) and is not concurrently under editorial evaluation elsewhere.

Integrity of Data, Algorithms, and Experimental Models

• Fabrication & Falsification: Empirical results, performance metrics, convergence curves, and data analytics must reflect genuine computational or experimental pipelines. Fabricating synthetic datasets without disclosure, selectively omitting outlying results to misrepresent model efficacy, or falsifying statistical significance constitutes severe research misconduct.
• Reproducibility & Data Availability: Authors are strongly encouraged to provide access to relevant source code, computational benchmarks, hyperparameter settings, and openly accessible repositories to verify the repeatability of computational intelligence models.
• Licensing and Authorized Assets: Authors must secure documented licensing rights for proprietary datasets, software libraries, third-party graphics, and architectural schemas incorporated within the manuscript prior to submission.
• Generative AI Ethics: In alignment with the journal’s policy, Generative AI (GenAI) may be utilized solely as auxiliary tools to assist with data processing, coding workflows, or language refinement. GenAI tools cannot be listed as an author or co-author. Authors bear complete accountability for factual correctness, mathematical veracity, citation reliability, and intellectual novelty.

Authorship Standards and Collective Accountability

• Authorship Criteria: Authorship must be limited strictly to individuals who made significant, substantive contributions to the conception, algorithmic architecture, implementation, empirical analysis, or drafting of the study. Administrative or general advisory roles do not qualify for authorship and should be recognized in the Acknowledgments.
• Corresponding Author Duties: The corresponding author serves as the liaison throughout submission, review, and publication, ensuring that all listed co-authors have reviewed, approved, and agreed to the final version of the manuscript.
• Dispute Resolution: Neither the editorial board nor the publisher accepts legal liability for intra-author disputes. Authorship conflicts must be resolved independently by the authors' host institutions.

3. Code of Ethics for Editors

Editorial Independence and Unbiased Decision-Making

The Editor-in-Chief and associate editorial officers hold absolute authority regarding manuscript acceptance, revision, or rejection based solely on technical rigor, intellectual depth, novelty, clarity, and thematic fit. The publisher maintains a strict firewall and does not intervene in editorial determinations.

All editorial appraisals are executed impartially, devoid of discrimination based on the author's nationality, race, institutional rank, gender, religious beliefs, or political perspective.

Double-Blind Peer Review Administration

• Double-Blind Integrity: Editors must ensure that author and reviewer identities remain strictly concealed from one another throughout the full peer review process.
• Expert Review Assignment: Every qualifying manuscript is assigned to a minimum of 3 independent reviewers possessing verified domain expertise in data science, artificial intelligence, machine learning, or allied computational disciplines.
• Finality of Decisions: Editorial determinations regarding scientific suitability, scope relevance, and peer review outcomes are final and binding. Editorial teams are insulated from external commercial, institutional, or personal interference.

Automated Similarity Screening & Plagiarism Thresholds

4. Code of Ethics for Reviewers

Review Timelines and Professional Obligations

Reviewers are requested to complete peer evaluations within an allocated window of 4 to 8 weeks. If an invited expert identifies a conflict of interest, lacks adequate domain specialization, or cannot fulfill the deadline, they must inform the editorial team immediately to allow prompt reassignment.

Reviewer assessments must be conducted personally. Delegating the review task to colleagues, postdoctoral researchers, or students without prior editorial consent is strictly prohibited.

Confidentiality, Intellectual Property, and Constructive Feedback

Manuscripts under review are privileged, confidential assets. Reviewers may not retain, duplicate, share, or appropriate proprietary algorithms, architectures, datasets, or theoretical findings for personal, commercial, or institutional research advantage.

Reviewers must provide objective, constructive, and evidence-based critiques focused on methodological validity, mathematical rigor, and clarity. Subjective, disparaging, or hostile commentary directed at authors is unacceptable.

5. Conflict of Interest (COI) Guidelines Matrix

Transparent identification of personal, professional, financial, or institutional biases is vital to preserving trust in computational research. The following matrix dictates the explicit obligations for all stakeholders:

Whistle-blower Protocols

If an undisclosed conflict of interest, systematic data manipulation, or ethical irregularity is brought forward by a whistle-blower, the editorial leadership will appoint an independent investigative panel. Confirmed violations will prompt immediate manuscript rejection or formal post-publication retraction in accordance with COPE frameworks.

6. Ethical Code for Studies Involving Human Data and Computational Ethics

Data Protection and Informed Consent

Where machine learning, computer vision, or data analytics models leverage personal records, clinical metrics, medical imaging, biometric templates, or human behavior profiles, authors must verify that documented, voluntary informed consent was secured prior to dataset compilation.

• Institutional Review Board (IRB) Clearance: Experimental protocols involving human subjects or proprietary personal records must provide formal documentation of ethics approval from an accredited Institutional Ethics Committee or Institutional Review Board.
• Vulnerable Cohorts: Where research encompasses datasets derived from minors or vulnerable groups, documented consent from legal guardians and strict anonymization protocols are mandatory.
• Publicly Available Data Exemptions: Studies exclusively analyzing fully open, non-restricted public datasets or synthetic data are exempt from specific consent requirements, provided the original repository licensing terms are honored.

Algorithmic Fairness, Privacy, and Social Impact

Given the wide deployment of computational intelligence in decision-support systems, manuscripts must consciously avoid reinforcing harmful societal biases or embedding discriminatory heuristics within algorithmic logic. Datasets must be thoroughly stripped of sensitive, personally identifiable attributes (e.g., telephone numbers, IP records, residential addresses) unless strictly essential and explicitly permitted by ethical oversight bodies.

7. Post-Publication Critiques, Corrections, and Retractions

Academic Discourse and Corrigenda

• Post-Publication Debates: JADSCI encourages rigorous academic commentary. Substantive methodological or mathematical critiques submitted to the editor will be shared with the authors for formal clarification. Both the critique and the authors' response may undergo peer assessment and subsequent publication.
• Errata & Corrections: When inadvertent mathematical, typographical, or computational errors are discovered that do not undermine the overall conclusions of the paper, an official Erratum or Corrigendum notice will be issued in the next scheduled issue.

Retraction Framework

In adherence to COPE Retraction Guidelines, JADSCI will retract a published article if:

• Definitive proof reveals that results are invalid due to severe computational errors, fabricated datasets, algorithmic miscalculations, or uncalibrated modeling pipelines.
• Unacknowledged plagiarism, image manipulation, or uncredited data copying exceeding accepted thresholds is confirmed post-publication.
• The manuscript breaches third-party intellectual property rights or was published without proper authorization.
• Unethical experimental methodologies or data harvesting breaches human participant protection laws.

Retraction Display: Retracted articles will remain accessible in the digital record to preserve scholarly continuity, but will be prominently stamped with a watermark indicating RETRACTED. A formal Retraction Note outlining the specific grounds will be appended permanently to the article's entry page.

Legal Indemnity

Intellectual disputes regarding dataset permissions or ownership must be resolved directly between the affected parties. The journal, its editors, reviewer panels, and publisher are legally indemnified against private authorship contestations, third-party copyright claims, or scope-related editorial rejections.

8. Open Access Statement, APC Policy, and Digital Archiving

Platinum Open Access & Zero-Fee Policy

The Journal of Applied Data Science and Computational Intelligence (JADSCI) operates as a fully Open Access journal under the Creative Commons Attribution 4.0 International License (CC BY 4.0). All published works are immediately and permanently free to read, download, redistribute, and build upon globally without subscription barriers.

No Article Processing Charges: JADSCI does not levy any submission fees, editorial processing costs, or Article Processing Charges (APCs) on authors. Publication costs are entirely underwritten to provide an equitable, high-caliber publishing forum for researchers worldwide.

Digital Preservation and Long-Term Accessibility

To ensure permanent digital persistence, manuscripts published in JADSCI are assigned permanent Digital Object Identifiers (DOIs) and preserved across open archival repositories including Zenodo and OpenAIRE. Dedicated repository snapshots protect against bit rot, server disruptions, and formatting deprecation over time.