SDM Workshop Recommender Systems vs RecSys

The SDM Workshop Recommender Systems Track and the ACM RecSys Conference differ in scope, size, and exam model: MLRec is a focused workshop co-located with SIAM Data Mining (SDM), while ACM RecSys is a fully standalone conference held since 2007. Choosing between them depends on whether you want a narrow topical venue or the field’s flagship stage.

Key Takeaways

• Two different professions. MLRec (the SDM workshop on machine learning and data mining for recommender systems) is a small thematic workshop; ACM RecSys is the field’s flagship comprehensive conference. They are complements and not substitutes.
• The review model is important. Workshops typically use lighter, quicker review with a program committee of specialists; ACM RecSys uses a multi-track, full-document review process with large PC and coaching.
• Calendar and location. SDM and its workshops move every year; ACM RecSys also runs every year. Check the official pages for the current year before planning submissions or trips.
• Publication Type. Workshop materials are generally short and unarchived or lightly archived; Full articles from ACM RecSys are archived and indexed in the ACM Digital Library.
• **Decide by objective. ** New and complete results → flagship conference. Start, position or applied work → workshop. Many researchers do both during their careers.

What Each Venue Actually Is

MLRec is a workshop on machine learning and data mining for recommender systems, co-located with the SIAM International Conference on Data Mining (SDM). SDM is organized under the SIAM Activity Group on Data Mining and is a long-running venue for data-mining methodology. The workshop format means a half-day to one-day program: a handful of accepted papers, invited talks, and discussion. The audience skews toward data-mining and ML researchers who care about the algorithmic and statistical side of recommendation.

ACM RecSys is the annual conference of the ACM Special Interest Group on Information Retrieval (SIGIR), dedicated entirely to recommender systems. It offers several tracks: full research papers, short articles, industrial track, reproducibility, doctoral conference, tutorials and workshops - and publishes the proceedings in the ACM digital library. Its scale is an order of magnitude larger than a single workshop.

The distinction is structural, not just prestige. A workshop is optimized for targeted exchange; a conference is optimized in terms of scale, archives and visibility on a field scale.

Comparison Table: MLRec (SDM Workshop) vs ACM RecSys

Dimension	MLRec (SDM workshop)	ACM RecSys
Type	Workshop co-located with SDM	Full standalone conference
Typical paper length	Short (workshop-length)	Full and short tracks
Review process	Workshop PC, lighter/faster	Multi-track, large PC, shepherding
Archival status	Often non-archival or lightly archived	Archival, ACM Digital Library
Program	Papers, invited talks, discussion	Papers, industry, tutorials, workshops, doctoral consortium
Best for	Early-stage, position, applied, focused ML/DM work	Complete, novel, high-impact results
Audience	ML/DM researchers, data miners	Full recommender-systems community
Frequency	Annual, tied to SDM’s location	Annual, rotating location

Use this table as a starting filter, then check the details against the current year’s call for papers, as formats and tracks evolve.

How the Two Venues Differ in Review and Acceptance

Depth of examination is the most marked practical difference. Full ACM RecSys articles go through a rigorous process with subject chairs, multiple reviewers, author response, and often a rebuttal phase; Accepted articles may be directed to address reviewers’ concerns. Workshop submissions at MLRec style venues generally receive less review from a smaller specialist committee, with decisions made more quickly.

Related: DataCamp — Browser-based, hands-on ML and data-science tracks you can start in 10 minutes.

Acceptance expectations differ accordingly. A workshop is designed to accept promising, incomplete, or discussion-oriented work that would struggle in a flagship journal. This is a feature, not a weakness: it’s where you pitch a half-formed idea and get expert feedback before committing to a full submission.

Archival implications are important for career planning. Archival conference papers can be cited, indexed, and considered for publication records in most evaluation frameworks. Unarchived workshop papers may not be, and some workshops explicitly allow work to be extended and resubmitted elsewhere - a useful property if you plan to turn a short article into a full article.

Program committee service also differs. Serving on a workshop PC is a low-commitment way to build review experience and visibility; Serving on the ACM RecSys PC is a broader, more structured commitment that carries weight in the community.

Our pick: Coursera (DeepLearning.AI & university specializations) — University- and industry-branded ML specializations with graded assignments and shareable certificates.

How to Decide Where to Submit

** Corresponds to the maturity of the work. **Comprehensive results with solid baselines, ablations and reproducibility artifacts are suitable for a flagship conference. Preliminary findings, negative results or a position on evaluation practice are suitable for a workshop.

Match the audience you need. If your contribution is a data mining or ML method (a novel factorization, a graph learning approach, a statistical estimator), the SDM workshop audience is precisely calibrated accordingly. If your contribution covers the recommender systems stack (interfaces, evaluation, industrial deployment, fairness), the broader conference reaches the right people.

Match your calendar. Workshop deadlines are typically later and lighter than conference deadlines, which is helpful if you have an outcome that isn’t yet conference ready but is worth discussing now.

Consider the two-step strategy. A common and effective route is to present early work at a workshop, gather feedback, and then submit the expanded version to the flagship conference. Check each venue’s policy regarding pre-presenting the workshop, as archiving status determines whether this is permitted.

Weigh visibility against fit. A flagship acceptance is more visible; a workshop acceptance is more likely and more focused. Neither is universally “better” — the right answer depends on what the paper needs next.

What the Recommender-Systems Venue Landscape Looks Like

ACM RecSys is the mainstay of the field, with its official website recsys.acm.org hosting current and past editions. SDM and its workshops fit into the broader data mining ecosystem alongside sites like KDD, WSDM, and CIKM, each with their own recommender systems content. Specialized Calendars, such as the community-managed conference calendar recommend-systems.com, track deadlines in these locations and are a convenient planning tool.

Related: Manning Publications — Deep, project-driven ML books and video courses — including the MEAP early-access program.

Adjacent communities also count. Information retrieval (SIGIR, ECIR), Web search and data mining (WSDM), and knowledge discovery (KDD) are all publishing work on recommender systems, and the boundaries are porous. An article on sequential recommendations may be suitable for RecSys, WSDM, or a data mining workshop depending on its framing.

Standards and reproducibility increasingly shape the choice of location. The ACM Artifact Review and Badging framework, along with reproducibility tracks at major conferences, rewards work with published code and data. If your contribution includes a strong artifact, a location with a reproducibility track amplifies its impact.

For authoritative background on the field, see the Wikipedia article on recommender systems and the ACM SIGIR organization page, which oversees ACM RecSys. SIAM’s activity group page for data mining describes the parent community behind SDM.

Where we would start: Udemy — One-off, low-cost ML and recommender-systems courses you own forever.

Practical Checklist Before You Commit

1. Read the current call for papers for both venues — tracks, lengths, and deadlines change yearly.
2. Confirm archival status if the paper needs to count as a publication or be resubmittable.
3. Check the review model — single vs. double-blind, number of reviewers, response period.
4. Assess audience fit by scanning recent accepted-paper titles from each venue.
5. Plan the extension path if you intend to grow a workshop paper into a conference paper.
6. Budget travel and time — co-located workshops add days to a conference trip.

Sources & Further Reading

• Recommender system — Wikipedia: A recommender system, also called a recommendation engine or content discovery platform is a type of information filtering system that aims to suggest items most…

Frequently Asked Questions

Is the SDM workshop on recommender systems the same as ACM RecSys?

No. MLRec is a focused workshop co-organized with the SIAM International Conference on Data Mining, while ACM RecSys is a fully standalone conference dedicated to recommender systems. They differ in size, rigor of review, archival status, and scope of program. Many researchers view them as complementary rather than competing places.

Which is better for a first paper — a workshop or a conference?

A workshop is usually the smoothest entry point: lighter revision, quicker decisions, and a specialized audience who give constructive feedback on early work. A flagship conference is suitable for comprehensive and well-validated results. If your work is preliminary or position-oriented, the workshop is the best first step.

Are workshop papers at SDM archival?

Archive status varies by workshop and year, so check the specific call for papers. Many workshops are not archived or are lightly archived, allowing authors to expand and resubmit their work at a later conference. ACM RecSys conference papers are archived and indexed in the ACM Digital Library.

How competitive is ACM RecSys compared to a workshop?

ACM RecSys is significantly more competitive because it is the field’s flagship venue with a large submission pool and a rigorous multi-track review process. Workshops accept a smaller, more focused set of articles with lighter editing. Competitiveness also depends on the specific track and year.

Can I submit the same paper to both a workshop and a conference?

It depends on the workshop’s archival policy and the conference’s prior-publication rules. Non-archival workshops often permit later conference submission of an extended version, but archival workshops may not. Always read both venues’ policies and disclose prior presentation where required.

Where can I find deadlines for these venues?

Visit the official ACM RecSys site at recsys.acm.org and the SIAM SDM site for the current year, as well as community-run calendars such as the recommendation-systems.com conference calendar. Deadlines change each year, so rely on primary sources rather than third-party summaries for final planning.

P.S. A few readers have asked which marketplace courses we actually reach for — it's Udemy; take a look here if you want the current details.