JUL192021_01B3203Decided 2021-07-19I-140

A university's petition to classify an assistant professor specializing in statistical machine learning as an…

Dismissed Useful for: avoid these mistakes
EB-1BField: statistical machine learning in business
The outcome

This appeal was not successful at this stage

The AAO dismissed the appeal because, although the Beneficiary met the minimum evidentiary criteria, the totality of the evidence did not establish that he is internationally recognized as outstanding in his academic field of statistical machine learning in business.

In plain English

The AAO dismissed the appeal filed by a university seeking EB-1B classification for an assistant professor in statistical machine learning. The Beneficiary satisfied three of the six evidentiary criteria—judging, original contributions, and scholarly articles—clearing the initial threshold. However, in the final merits determination the AAO found that citation counts were modest, expert letters lacked corroborating detail, peer review participation was routine rather than indicative of elite standing, and media coverage described incomplete work. The AAO also rejected the Clarivate Analytics citation comparison as methodologically flawed. The decision underscores that meeting the minimum evidentiary criteria does not guarantee approval; the totality of evidence must demonstrate genuine international recognition as outstanding.

What worked & what failed

What worked: The Beneficiary met three criteria: judging the work of others through peer review for conferences and journals, making original research contributions in healthcare machine learning, and authoring scholarly articles. These satisfied the initial evidentiary threshold.

What failed: Citation counts were low and the Clarivate comparison was methodologically unreliable. Expert letters relied on conclusory language without concrete examples of broad field impact. The 'Junior Researcher' Best Paper Award was not shown to carry international prestige. Media coverage in New Scientist and other outlets noted the algorithm still needed significant work, undermining claims of outstanding achievement.

Takeaway: For EB-1B petitions, meeting the two-criterion threshold is just the starting point—petitioners must build a compelling narrative of international recognition through high citation counts (with self-citation analysis), prestigious editorial board roles, widely adopted contributions, and expert letters that provide specific, concrete evidence of field-wide impact rather than restating statutory language.

For RFE responses & petition building

Cases like this are frequently used by attorneys when responding to RFEs or building initial petitions. The evidence patterns that worked (or failed) here directly reflect what USCIS officers look for when evaluating EB-1B criteria.

Evidence that moved the needle

  • The Beneficiary met three criteria: judging the work of others through peer review for conferences and journals, making original research contributions in healthcare machine learning, and authoring scholarly articles
  • These satisfied the initial evidentiary threshold.

Evidence that wasn't enough alone

  • Citation counts were low and the Clarivate comparison was methodologically unreliable
  • Expert letters relied on conclusory language without concrete examples of broad field impact
  • The 'Junior Researcher' Best Paper Award was not shown to carry international prestige
  • Media coverage in New Scientist and other outlets noted the algorithm still needed significant work, undermining claims of outstanding achievement.
Find more EB-1B cases with similar evidence patterns →
What the evidence showed

Criterion-by-criterion breakdown

Lesser nationally or internationally recognized prizes or awards

Not met

Beneficiary received a 'Junior Researcher' Best Paper Award but it was not shown to be commensurate with major prizes for outstanding achievement at the international level.

Published material about the person

Not met

Articles in New Scientist, Netzwoche, Pressetext, and The Economist did not demonstrate international recognition as outstanding; articles noted the algorithm still needed work.

Judging the work of others

Met

Director and AAO agreed criterion was met; however at final merits the peer review activity was found insufficient to show international recognition as outstanding.

Original contributions of major significance

Met

Director and AAO agreed criterion was met; however expert letters lacked corroborating evidence showing wide influence or international recognition at the outstanding level.

Authorship of scholarly articles

Met

Director and AAO agreed criterion was met; however citation counts were modest and Clarivate Analytics comparison was methodologically flawed.

Evidence that persuaded the AAO

Peer review activities for multiple conferences and journals (ICIS, CIST, INFORMS Data Science Workshop, UAI, CHITA Best Paper Competition, MISQ, ISR, Management Science, Decision Sciences Journal)

Original research contributions in statistical machine learning and healthcare applications

Authorship of approximately 17 scholarly articles with top four articles receiving 25, 21, 10, and 9 citations respectively

Three evidentiary criteria satisfied: judging, original contributions, and scholarly articles

Where the evidence fell short

Expert recommendation letters found to merely repeat statutory language without sufficient corroborating detail showing widespread influence

Clarivate Analytics citation percentile data found methodologically flawed due to non-contemporaneous comparison and unexplained field selection

New Scientist, Netzwoche, and Pressetext articles noted the algorithm still needed work and did not establish outstanding international recognition

The Economist article not about the Beneficiary's work and including only one sentence mentioning him

Wikipedia-sourced information about Netzwoche circulation deemed unreliable

Pressetext self-promotional publisher material not accepted

Post-filing peer review activities (multiple emails dated after petition filing) not counted

'Junior Researcher' Best Paper Award not shown to be a major prize of international significance

Program co-chair role at 2020 workshop post-dated petition filing and could not be considered

How the case moved

Completed

I-140 filed

Assistant Professor of Economics specializing in statistical machine learning, information systems, and analytics

Completed

Director — Denied

Initial decision: Denied.

Completed

Appeal to the AAO

Petitioner appealed to the Administrative Appeals Office for de novo review.

2021-07-19

AAO decision — Dismissed

The AAO dismissed the appeal because, although the Beneficiary met the minimum evidentiary criteria, the totality of the evidence did not establish that he is internationally recognized as outstanding in his academic field of statistical machine learning in business.

If you're appealing a similar decision, I-290B must be filed within 30 days of personal service of the denial, or 33 days if mailed.

Find this useful? A coffee helps keep Case Reviewer free and ad-free.

Buy me a coffee
Authorities the office relied on
8 C.F.R. § 204.5(i)(3)(i)8 C.F.R. § 204.5(i)(3)(i)(A)8 C.F.R. § 204.5(i)(3)(i)(C)8 C.F.R. § 204.5(i)(3)(i)(A)-(F)8 C.F.R. § 204.5(i)(3)(ii)8 C.F.R. § 103.2(b)(1)8 C.F.R. § 103.2(b)(12)
ChawathePreponderance of the evidence standard; petitioner must show eligibility is more likely than not true.
E-M-Quality and quantity of evidence both considered in preponderance analysis.
Fedin Bros.Merely repeating statutory or regulatory language does not satisfy the petitioner's burden of proof.
BadasaWikipedia cannot guarantee the validity of its content and is not a reliable evidentiary source.
Braga v. PoulosUSCIS need not rely on self-promotional material of a publisher.