mlsanityData to 7 Oct 2026

Data Smells in Public Datasets

arXiv 2203.08007 ↗First posted 15 Mar 2022cs.SE cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

The adoption of Artificial Intelligence (AI) in high-stakes domains such as healthcare, wildlife preservation, autonomous driving and criminal justice system calls for a data-centric approach to AI. Data scientists spend the majority of their time studying and wrangling the data, yet tools to aid them with data analysis are lacking. This study identifies the recurrent data quality issues in public datasets. Analogous to code smells, we introduce a novel catalogue of data smells that can be used to indicate early signs of problems or technical debt in machine learning systems. To understand the prevalence of data quality issues in datasets, we analyse 25 public datasets and identify 14 data smells.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 11 of its 61 references that are in the map. First posted on arXiv 15 Mar 2022.

Builds on

11 of 61 references.

References · 11

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019Machine Learning Testing: Survey, Landscapes and Horizonscited by 213
  3. 2020Towards Accountability for Machine Learning Datasets: Practices from Software Engineering and Infrastructurecited by 138
  4. 2021Trustworthy AI: A Computational Perspectivecited by 93
  5. 2018Software Engineering Challenges of Deep Learningcited by 41
  6. 2020Engineering AI Systems: A Research Agendacited by 37
  7. 2020AI Lifecycle Models Need To Be Revised. An Exploratory Study in Fintechcited by 33
  8. 2017BoostClean: Automated Error Detection and Repair for Machine Learningcited by 23
  9. 2021Becoming Good at AI for Goodcited by 8
  10. 2021On the experiences of adopting automated data validation in an industrial machine learning projectcited by 6
  11. 2022On the Prevalence, Impact, and Evolution of SQL Code Smells in Data-Intensive Systemscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 0; 2024: 2; 2025: 0; 2026: 2.

5 papers.

  1. 2024LLM-based feature generation from text for interpretable machine learningcited by 12
  2. 2022Towards Using Data-Influence Methods to Detect Noisy Samples in Source Code Corporacited by 4
  3. 2024What About the Data? A Mapping Study on Data Engineering for AI Systemscited by 2
  4. 2026Unmasking LAION-5B: Age, Gender, Race, and Emotion Biases in Large-Scale Image Datasetscited by 0
  5. 2026From AI Technical Debt to Agentic Technical Debt: A Systematic Mapping of Root Causes and Manifestations in Agentic AI Systemscited by 0