Continuous Submissions

Continuous submission. Manuscripts can be submitted at any time from the opening date (see Time Scale) until 31 January 2027, while intake is open. Each is processed on arrival.

Action editors. Every manuscript is handled by an action editor (AE), who reads it, decides (together with Journal Track chairs) whether it goes out for review, and oversees the reviewing and revisions. The AEs are listed under the JT Programme Committee.

Intake cap. The number of manuscripts sent out for review per journal (MLJ and DAMI separately) is capped by the number of action editors onboarded. The cap guarantees that every manuscript sent out for review receives prompt, high-quality review. The cap is raised as more action editors join; our target is at least 100 manuscripts per journal. Places are filled first come, first served. When a journal's cap is reached, the Journal Track is closed for that journal, and no further manuscripts can be sent out for review until the cap is raised. Intake closes on 31 January 2027 in any case.

Given the special nature of the Journal Track, submitted papers must adhere to the following eligibility criteria:

• Papers must not be under review elsewhere at any time between submission and notification.
• Papers must satisfy the high-quality criteria of the journals involved and, at the same time, lend themselves to conference talks.
• Journal versions of previously published conference papers and survey papers will not be considered for the Journal Track.
• Papers rejected in the ECML PKDD 2027 Journal Track may not be resubmitted to it, regardless of modifications. Additionally, papers rejected by the ECML PKDD Journal Track of one journal cannot be submitted to the ECML PKDD Journal Track of the other journal.
• Papers must meet the reproducible research requirement below.
• Papers that do not fall into the eligible category may be rejected without formal reviews. As usual in journal review processes, papers may be desk rejected at the discretion of the editor-in-chief and/or other editors in charge of the submission.
• Authors who submit their work to the ECML PKDD 2027 Journal Track commit themselves to registering and presenting their paper at ECML PKDD 2027 if it is accepted by 20 June 2027.

The Journal Track requires reproducible research. Authors make the data, software and documentation needed to reproduce the reported results available, or explain in the cover letter why this is not possible.

At submission: the cover letter gives a link to the material, or a stated reason why it cannot be shared. Acceptable reasons include confidential data and privacy or legal restrictions. Theoretical papers without experiments answer "not applicable"; their proofs and other claims must be detailed enough to be verified, and the main text must convey the main results and the key ideas of the proofs.

Partial sharing: confidential data or algorithms are not a reason to share nothing. Where possible, authors should still provide a working implementation, for example with anonymised or toy data.

Desk rejection: a submission with neither the material nor a justification may be desk rejected.

Upon acceptance: the material must be archived with a persistent identifier, for example a DOI from Zenodo (which can archive a GitHub release directly) for data and code, or a Software Heritage identifier (SWHID) for source code; a link to GitHub or GitLab alone is not persistent. The identifier must be cited in the final version of the paper.

Manuscripts submitted to the Journal Track that receive the final acceptance decision by 20 June 2027 will be presented at ECML PKDD 2027. Papers accepted later may be presented at ECML PKDD 2028. Submitted papers that are rejected in the Journal Track may be recommended, at the discretion of the Journal Track Chairs, for submission to the Findings of ECML PKDD track (more details on this track will follow).

The Journal Track accepts continuous submissions from the opening date until 31 January 2027; all dates are AoE. Submissions open as soon as the submission systems are ready, and at the latest on 30 November 2026; the opening date will be announced on this page in advance. Submissions after 31 January 2027 are not considered for ECML PKDD 2027. Arrangements for submissions after that date, if any, may be announced later.

We strive for a high-quality and efficient review process. Our targets are a first decision about 8 weeks after submission, and acceptance within about 4.5 months for papers that require only minor revisions, though meeting these targets may not always be possible. Papers not rejected in the initial review phase normally require (sometimes substantial) revisions, and the revised paper will be re-reviewed, which extends the review process. For reference, in 2026, every accepted paper went through at least two rounds of revision. Late submissions that require major revisions will therefore probably miss the cut-off for presentation at ECML PKDD 2027. We therefore advise submitting as early as possible, but not before the paper is ready.

The reviewing process is single-blind (so the authors' identities are disclosed to the anonymous reviewers). Each submission is reviewed following the policies of the journal concerned.

To facilitate both reviewer allocation and journal-style reviewing, authors submitting to the ECML PKDD 2027 Journal Track must complete two submission processes. Each manuscript must be submitted both to OpenReview and to the relevant journal's dedicated SNAPP submission platform. OpenReview will be used solely for reviewer allocation and conflict-of-interest management, while the full review and editorial process will take place entirely within the journal's system. Specifically:

• Submit the title, abstract, the cover letter and manuscript PDF (the same version as submitted to SNAPP) to OpenReview, adding the author names and relevant keywords and indicating which of the two journals the paper is intended for. All authors need an OpenReview profile. Approving new profiles can take time, so please create yours well in advance.
• Submit the full paper to SNAPP (selected journal only), choosing the collection ECML PKDD 2027 during the submission: either to the Springer Data Mining and Knowledge Discovery journal, or to the Springer Machine Learning journal.

Both submissions must be completed while intake is open; a submission is processed and takes its place under the cap once both are complete. No paper can be submitted to both journals. Each paper has exactly one corresponding author, who is determined upon submission and cannot be changed later. Check all responsibilities of the corresponding author carefully at https://support.springernature.com/en/support/solutions/articles/6000214118-first-author-and-corresponding-author-defined

Incomplete submissions (missing either the submission in the SNAPP system of the selected journal or the submission in OpenReview), as well as double submissions to both journals, will be desk rejected. Papers MUST be prepared in English and formatted following the submission guidelines on the respective journals' home pages (Submission guidelines for Data Mining and Knowledge Discovery, Submission guidelines for Machine Learning). Templates can be found at https://www.springernature.com/gp/authors/campaigns/latex-author-support

Submitted papers must not exceed 21 pages, including references. Papers exceeding this limit may be desk rejected. Every paper may be accompanied by unlimited appendices (as part of the same file). Reviewers are not required to read the appendices, so the main text must be self-contained and contain everything needed to assess the main claims. Important content that appears only in the appendices is grounds for rejection. Reviewers may, at their discretion, consult the appendices, for example to check detailed proofs.

Cover letter. Both journals require authors to include an information sheet (for Machine Learning submissions) or a cover letter (up to 2 pages) as supplementary material (for Data Mining and Knowledge Discovery submissions) that contains a short summary of their contribution and specifically addresses the following questions:

• What is the main claim of the paper? Why is this an important contribution to the machine learning/data mining research area?
• What is the evidence provided to support claims? Be precise.
• Report 3–5 most closely related contributions in the past 7 years (authored by researchers outside the authors' research group) and briefly state the relation of the submission to them.
• Specify 5 general keywords and 5 specific keywords describing the main research activity presented in the manuscript.
• Where are the data and code needed to reproduce the results available? If they cannot be shared, why not? (See Reproducible Research; theoretical papers answer "not applicable".) Answering this question does not replace the other requirements in Reproducible Research.
• Declare any conflict of interest by reporting the email domains of all institutions with which the authors have an institutional conflict of interest. Authors have an institutional conflict of interest if they are currently employed or have been employed at this institution in the past three years, or if the authors have extensively collaborated with this institution within the past three years. Authors are also required to identify all action editors with whom the authors have a conflict of interest. Examples of conflicts of interest include co-authorship in the last five years, a colleague in the same institution within the last five years, advisor/student relationships, parentage relationships.
• Suggest the most suitable action editors for the paper and suggest potential reviewers. Who are the most appropriate reviewers for the paper? Authors are required to suggest up to four candidate reviewers, including a brief motivation for each suggestion. Optionally, list up to four researchers/potential reviewers with competing interests that should not be considered for reviewers. PhD students cannot be suggested as potential reviewers.

Research reported at ECML PKDD should avoid harm, be honest and trustworthy, fair and non-discriminatory, and respect privacy and intellectual property. Where relevant, authors can include in the main body of their paper, or on the reference page, a short ethics statement that addresses ethical issues regarding the research being reported and the broader ethical impact of the work. Reviewers will be asked to flag possible violations of relevant ethical principles. Such flagged submissions will be reviewed by the editors-in-chief, Journal Track chairs, or action editors. Authors may be required to revise their paper to include a discussion of possible ethical concerns and their mitigation.

In particular, ECML PKDD subscribes to the European Code of Conduct for Research Integrity. By submitting to ECML PKDD, you confirm that you are aware of and accept the following specific policies:

• Originality: Submissions must not have substantial overlap in either contribution or text with work previously accepted for publication as a full paper in another archival forum. Papers at workshops without archival proceedings and preprints are fine.
• Concurrent submission: We recognise the significant strain on the scientific community caused by the needs of peer review. Therefore, the work you submit must not be under review elsewhere at any time between submission and notification.
• Single-blind review: Reviewing for the ECML PKDD Journal Track is single-blind, meaning that the identities of the reviewers are not disclosed to the authors.
• Reproducibility: The Journal Track requires reproducible research (see Reproducible Research). Reviewers will be instructed to pay close attention to the reproducibility of results where appropriate.
• Information sharing: Submissions will be treated confidentially. However, papers, author information, and reviews may be shared with the organisers of other AI conferences to identify duplicate submissions and to limit duplicate reviewing efforts.
• Abstracts: Abstracts are central for the assignment of reviewers. Therefore, they must not be altered in any significant way after submission. In particular, submitting "placeholder abstracts" is not admissible.
• Authorship: All individuals, and only those, who have made significant contributions to a paper should be listed as authors in the submission system. We will not permit adding or removing authors to a paper after submission. Please also ensure that the order of authors is correct at submission time.
• Reviewing: To ensure that all papers receive a sufficient number of high-quality reviews, there is a requirement for authors to contribute to reviewing. Every submission must nominate at least one author who is a qualified reviewer (i.e., authors with at least two papers in ECML PKDD or other top-ranked related conferences or journals). If no qualified reviewer appears on the author list, the authors should nominate the best-qualified author for consideration by the Journal Track chairs. In addition, any author listed on 3 or more papers will be automatically signed up as a reviewer unless they are already serving as a reviewer, action editor, AC, or SAC or they are not sufficiently qualified to serve on the PC (such as students). Either case above constitutes an acceptance and a commitment to carry out the regular reviewing load responsibly. Failure to provide a qualified reviewer when one exists in the author list, or failure to carry out the assigned reviewing duty properly, is grounds for desk rejection.

Journal Track chairs: Devendra Singh Dhami, Antonio Vergari, Kai Puolamäki

Action Editors: (to be added)