Scholar4AI Updated 2026-11-03 18:03 UTC+09:00

Papers on “outliers”

Research on “outliers” in the 480 tracked journals: papers per year and category, the journals, institutions and countries publishing it, the keywords studied with it and its most-cited papers. It is not among the most frequent keywords of the last 30 days, so this page has no list of its latest papers. Also as Markdown.

Share: X LinkedIn Bluesky Threads Facebook Reddit Hacker News Hatena LINE Email
Copy:

← Top research keywords · Keyword trends

Key findings#

  • From 2017 to 2026 (so far), the keyword “outliers” appeared in 477 papers in the tracked journals, most of them in statistics (225), AI (147) and operations research (47).
  • 66 of them were published in 2025, against 36 in 2017.
  • Since 2023, the institutions with the most of these papers were Beijing Normal University (5), Huazhong Agricultural University (5) and Indian Statistical Institute (5).
  • The journals publishing most of them were Statistics and Computing (23), Statistical Papers (14) and Computational Statistics & Data Analysis (13).
  • The most-cited of them is “A comprehensive survey of loss functions and metrics in deep learning” (Artificial Intelligence Review, 2025), with 401 citations.

Papers per year#

2017: 36 papers2018: 50 papers2019: 35 papers2020: 59 papers2021: 53 papers2022: 41 papers2023: 47 papers2024: 49 papers2025: 66 papers2026 (so far): 41 papers
20172026 (so far)
Papers per publication year and category
Category2017201820192020202120222023202420252026 (so far)
Statistics19161731152221273027
AI922101422101813218
Operations Research3536753483
Economics3433724342
Finance1223211021
Higher Education and Research0101010000
Management0001000100
Health Economics and Policy0000000010
Accounting1000000000
Information Systems0000000100
All categories36503559534147496641

Where the papers come from#

Institutions with the most papers, since 2023
#InstitutionPapers
1Beijing Normal University5
2Huazhong Agricultural University5
3Indian Statistical Institute5
4The University of Texas at El Paso5
5Beihang University4
6Australian National University3
7Beijing Normal University, Zhuhai3
8Chiba University3
9Consejo Nacional de Investigaciones Científicas y Técnicas3
10Henan University3
Countries with the most papers, since 2023
#CountryPapers
1China81
2United States30
3Australia12
4India12
5Canada11
6Turkey10
7Italy9
8South Korea9
9Japan8
10Hong Kong6
Keywords with the most papers, since 2023
#KeywordPapers
1robust estimation54
2robust regression21
3heavy-tailed errors17
4heavy-tailed distributions15
5high-dimensional data15
6influence function15
7contaminated data14
8feature selection12
9asymptotic normality10

Most-cited papers#

Most-cited papers since 2017
#CitationsPer monthTitleJournalPublishedAuthors
140121.4A comprehensive survey of loss functions and metrics in deep learningArtificial Intelligence Review2025-04-11Juan R. Terven, Diana Margarita Córdova Esparza, Julio-Alejandro Romero-González et al.
23403.4A generalized mean distance-based k-nearest neighbor classifierExpert Systems with Applications2018-08-13Jianping Gou, Hongxing Ma, Weihua Ou et al.
32092.0Mixture correntropy for robust learningPattern Recognition2018-02-21Badong Chen, Xin Wang, Na Lü et al.
41922.4Radial-Based Undersampling for imbalanced data classificationPattern Recognition2020-02-06Michał Koziarski
51381.4The impact of globalization and economic freedom on economic growth: the case of the Latin America and Caribbean countriesEconomic Change and Restructuring2018-11-12Renato Santiago, José Alberto Fuinhas, António Cardoso Marques
Cite this page

Scholar4AI. (2026). Papers on “outliers”. Journal and Conference Research Rankings. Retrieved November 3, 2026, from https://scholar4ai.com/keywords/outliers

@misc{scholar4ai-keywords-outliers,
  author = {{Scholar4AI}},
  title = {Papers on “outliers”},
  year = {2026},
  howpublished = {\url{https://scholar4ai.com/keywords/outliers}},
  urldate = {2026-11-03},
  note = {Journal and Conference Research Rankings; data from OpenAlex (CC0); licensed CC BY-NC 4.0}
}

Free to reuse for non-commercial purposes with attribution (CC BY-NC 4.0).