Learn2Clean: Optimizing the Sequence of Tasks for Web Data Preparation
Open accessThe DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Data cleaning and preparation has been a long-standing challenge in data science to avoid incorrect results and misleading conclusions obtained from dirty data. For a given dataset and a given machine learning-based task, a plethora of data preprocessing techniques and alternative data curation strategies may lead to dramatically different outputs with unequal quality performance. Most current work on data cleaning and automated machine learning, however, focus on developing either cleaning algorithms or user-guided systems or
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The first of these relates to data iteration, where the steps of model building and testing are frequently followed by the need to collect new data, or enriching the existing dataset [5,11,31,34].”
From A Survey of Data Quality Requirements That Matter in ML Development Pipelines (2023) · cites this paper as [11]
In the citing paper
“As mentioned during the pre-processing stage, automated techniques can be used to select the optimal sequence of data preprocessing tasks that maximise the performance of the ML model [11].”
From A Survey of Data Quality Requirements That Matter in ML Development Pipelines (2023) · cites this paper as [11]
Citations
Cites
mice: Multivariate Imputation by Chained Equations in R
Buuren, S. V., & Groothuis-Oudshoorn, K. (2011) · Journal of Statistical Software · doi:10.18637/jss.v045.i03 · Cited by 13,870 · cited 1×
“Four imputation methods were used to im-pute missing values: two distance-based imputation methods: Expectation-Maximization (EM), K-NN [1], Multiple Imputation by Chained Equations ( MICE ) [5], and replacement by the most frequent value (MF); • Outlier detection .”
MethodWrappers for feature subset selection
Kohavi, R., & John, G. H. (1997) · Artificial Intelligence · doi:10.1016/s0004-3702(97)00043-x · Cited by 8,947 · cited 1×
“Four methods were tested to select an optimal set of features, namely: based on a user-defined acceptable ratio of missing values (MR), removing collinear features (LC), using a wrapper subset evaluator (WR) [8], and a tree-based classifier for feature selection (TB); • Imputation .”
MethodWatkins, C. (1989) · OpenGrey (Institut de l'Information Scientifique et Technique) · doi:10.17863/cam.132164 · Cited by 5,297
LOF: identifying density-based local outliers
Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000) · ACM SIGMOD Record · doi:10.1145/335191.335388 · Cited by 5,207
(2015) · neural information processing systems · Cited by 1,305
OpenML: networked science in machine learning
Vanschoren, J., van Rijn, J. N., Bischl, B., & Torgo, L. (2014) · ACM SIGKDD Explorations Newsletter · doi:10.1145/2641190.2641198 · Cited by 1,141
Data Cleaning: Overview and Emerging Challenges
Chu, X., Ilyas, I. F., Krishnan, S., & Wang, J. (2016) · Proceedings of the 2016 International Conference on Management of Data · doi:10.1145/2882903.2912574 · Cited by 488 · cited 1×
“…the information presented on Web There are several methods that optimize single data cleaning operations based on downstream accuracy metrics (see [18] for a survey) and also recent approaches such as [17], data transformation learning [22], and AutoML line of work [19, 20] in the area of…”
MethodMultivariate outlier detection in exploration geochemistry
Filzmoser, P., Garrett, R. G., & Reimann, C. (2005) · Computers & Geosciences · doi:10.1016/j.cageo.2004.11.013 · Cited by 460 · cited 1×
“Outliers were detected using three meth-ods including: a statistic-based approach, Inter Quartile Range (IQR), the Z-score-based method (ZSB) [7], and a density-based approach (Local Outlier Factor, LOF) [4].”
MethodA Study of K-Nearest Neighbour as an Imputation Method.
Batista, G. E. A. P. A., & Monard, M. C. (2002) · Cited by 329
Discovery of complex glitch patterns: A novel approach to Quantitative Data Cleaning
Berti-Equille, L., Dasu, T., & Srivastava, D. (2011) · 2011 IEEE 27th International Conference on Data Engineering · doi:10.1109/icde.2011.5767864 · Cited by 63 · cited 2×
“In particular, we plan further experiments with semi-synthetic data where noise injection is carefully controlled in order to study more complex and intricate distributions of various glitch types and patterns [3] (combining duplicates, inconsistencies, missing and outlying values) and their impact on the robustness o…”
BackgroundMethod
Research works printed but not linked: 3
- Berti-Equille L. , Scannapieco M. ( 2016 ). Quality of Web Data (Chapter). In the 2 nd Edition of the book Data Quality : Concepts, Methodologies and Techniques, Springer , 2016. Berti-Equille L., Scannapieco M. (2016). Quality of Web Data (Chapter). In the 2nd Edition of the book Data Quality: Concepts, Methodologies and Techniques, Springer, 2016.
- Bilalli B. , Abelló A. ( 2018 ). PRESISTANT: Learning based assistant for data preprocessing , Arxiv CoRR abs/1803.01024. Bilalli B., Abelló A. (2018). PRESISTANT: Learning based assistant for data preprocessing, Arxiv CoRR abs/1803.01024. · arXiv
- Feurer M. , Hutter F. ( 2018 ). Towards Further Automation in AutoML , Proc. of ICML 2018 AutoML Workshop. Feurer M., Hutter F. (2018). Towards Further Automation in AutoML, Proc. of ICML 2018 AutoML Workshop.
Cited by
Overview and Importance of Data Quality for Machine Learning Tasks
Jain, A., Patel, H., Nagalapatti, L., Gupta, N., Mehta, S., Guttula, S., Mujumdar, S., Afzal, S., Sharma Mittal, R., & Munigala, V. (2020) · Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining · doi:10.1145/3394486.3406477 · Cited by 293
“We further plan to discuss few automated data cleaning approaches proposed in the literature [1] and finally end the topic by illustrating a real life use-case of deployed data cleaning system in Google [2] and its benefits.”
MethodData Quality for Machine Learning Tasks
Gupta, N., Mujumdar, S., Patel, H., Masuda, S., Panwar, N., Bandyopadhyay, S., Mehta, S., Guttula, S., Afzal, S., Sharma Mittal, R., et al. (2021) · Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining · doi:10.1145/3447548.3470817 · Cited by 127
“This raises some interesting questions: what should be the guiding factors by which such a sequence can be chosen, how an we quantitatively measure these factors, and, is it possible to find an optimal sequence that allows us to maximize the ML classifier performance [5].”
BackgroundData cleaning and machine learning: a systematic literature review
Côté, P.-O., Nikanjam, A., Ahmed, N., Humeniuk, D. V., & Khomh, F. (2024) · Automated Software Engineering · doi:10.1007/s10515-024-00453-w · Cited by 163
“For example, Berti-Equille (2019) designed an RL agent that learns to clean a dataset using a set of data cleaning tools that can clean different types of errors.”
BackgroundMethodInfluentialA Survey of Data Quality Requirements That Matter in ML Development Pipelines
Priestley, M., O’donnell, F., & Simperl, E. (2023) · Journal of Data and Information Quality · doi:10.1145/3592616 · Cited by 86
“The first of these relates to data iteration, where the steps of model building and testing are frequently followed by the need to collect new data, or enriching the existing dataset [5, 11, 31, 34].”
BackgroundMethodInfluentialCollaborative Edge Server Placement for Maximizing QoS With Distributed Data Cleaning
Liang, Y., Yin, M., Wang, W., Liu, Q., Wang, L., Zheng, X., & Wang, T. (2025) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2025.3552337 · Cited by 35
“(cid:2) Learn2clean [30]: This strategy employs a model-independent reinforcement learning approach designed to determine the most effective sequence of preprocessing tasks for a specific dataset, machine learning model, and quality performance metric.”
Hicham, N., & Karim, S. (2022) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2022.0131016 · Cited by 27
“1) Data preparation: Data selection, preprocessing, and transformation are the three stages of the process that are involved in getting data suitable for an algorithm that does machine learning [20] [21].”
BackgroundDiffPrep: Differentiable Data Preprocessing Pipeline Search for Learning over Tabular Data
Li, P., Chen, Z., Chu, X., & Rong, K. (2023) · Proceedings of the ACM on Management of Data · doi:10.1145/3589328 · Cited by 57
“These datasets are frequently used in the AutoML and data cleaning literature [5, 20, 23, 24, 28].”
MethodInfluentialAdvances in Exploratory Data Analysis, Visualisation and Quality for Data Centric AI Systems
Patel, H., Guttula, S., Mittal, R. S., Manwani, N., Berti‐Équille, L., & Manatkar, A. (2022) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3534678.3542604 · Cited by 26
“We will also discuss the challenges associated with running advanced data quality metrics for ML as described in [10], automated sequencing [4] as well as performing EDA and visualisation tasks on large scale datasets.”
MethodHAIPipe: Combining Human-generated and Machine-generated Pipelines for Data Preparation
Chen, S., Tang, N., Fan, J., Yan, X., Chai, C., Li, G., & Du, X. (2023) · Proceedings of the ACM on Management of Data · doi:10.1145/3588945 · Cited by 26
“Learn2clean [2] and its extension [3] are pioneering works of using reinforcement learning for generating data prep pipelines for ML.”
BackgroundSAGA: A Scalable Framework for Optimizing Data Cleaning Pipelines for Machine Learning Applications
Siddiqi, S., Kern, R., & Böehm, M. (2023) · Proceedings of the ACM on Management of Data · doi:10.1145/3617338 · Cited by 49
“Systems such as ActiveClean [58], BoostClean [57], AlphaClean [59], CPClean [53], Learn2Clean [8] already use these signals, but focus on specific models or types of errors (e.g., missing values).”
MethodRelational Data Cleaning Meets Artificial Intelligence: A Survey
Zhu, J., Zhao, X., Sun, Y., Song, S., & Yuan, X. (2024) · Data Science and Engineering · doi:10.1007/s41019-024-00266-7 · Cited by 41
“Learn2Clean [115], based on the model-free reinforcement learning technique Q-Learning, devises the optimal task sequence for pre-processing, given a dataset, ML model, and performance metrics, thereby maximizing the quality of the ML model.”
Data pre-processing pipeline generation for AutoETL
Giovanelli, J., Bilalli, B., & Abelló, A. (2021) · Information Systems · doi:10.1016/j.is.2021.101957 · Cited by 26
Enhancing data preparation: insights from a time series case study
Sancricca, C., Siracusa, G., & Cappiello, C. (2024) · Journal of Intelligent Information Systems · doi:10.1007/s10844-024-00867-8 · Cited by 21
“…both tabular and streaming data); (ii) our approach covers the majority of the data preparation actions and data analysis tasks (only Learn2clean (Berti-Équille, 2019, 2020) is comparable to our approach); finally, (iii) DIANA is one of the few human-centered approaches (only (Martin et al.,…”
Automatic Assessment of Quality of your Data for AI
Patel, H., Gupta, N., Panwar, N., Mittal, R. S., Mehta, S., Guttula, S., Mujumdar, S., Afzal, S., Bedathur, S., & Munigala, V. (2022) · Joint International Conference on Data Science & Management of Data (CODS-COMAD) · doi:10.1145/3493700.3493774 · Cited by 18
The matching scarcity problem: When recommenders do not connect the edges in recruitment services
Cardoso, A., Mourão, F., & da Rocha, L. C. D. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.114764 · Cited by 14
DQLearn : A Toolkit for Structured Data Quality Learning
Shrivastava, S., Patel, D., Zhou, N., Iyengar, A., & Bhamidipaty, A. (2020) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata50022.2020.9378296 · Cited by 18
“This approach is motivated by work done before [21] [22] [26], however its integration with Workflow Graph is unique.”
MethodReinforcement Learning for Data Preparation with Active Reward Learning
Berti‐Équille, L. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-34770-3_10 · Cited by 16
Machine Learning Applications for Consumer Behavior Prediction
Hicham, N., & Karim, S. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-26852-6_62 · Cited by 15
Sustainable Quality in Data Preparation
Pernici, B., Cappiello, C., Bono, C. A., Sancricca, C., Catarci, T., Angelini, M., Filosa, M., Palmonari, M., Paoli, F. D., Bergamaschi, S., et al. (2025) · Journal of Data and Information Quality · doi:10.1145/3769120 · Cited by 13
“As regards data cleaning operations, [94] automatically generates data repairs through a probabilistic approach; in [19, 20], the design of a data preparation pipeline is driven by the maximization of the quality of the results of an ML model.”
Collaborative Edge Service Placement for Maximizing QoS with Distributed Data Cleaning
Liang, Y., Wang, W., Zheng, X., Liu, Q., Wang, L., & Wang, T. (2023) · International Workshop on Quality of Service · doi:10.1109/iwqos57198.2023.10188694 · Cited by 10
“3) Learn2clean [13]: a model-free reinforcement learning strategy that determines the ideal sequence of tasks for preprocessing the data to optimize the quality of the ML model output.”
MethodA Survey on Data Asset Value Change Estimation and Appreciation with Data Governance
Ding, X., Li, G., Tang, Y., Liang, C., Yu, T., Zhou, M., Liu, Y., Qian, Z., Song, Z., & Wang, H. (2025) · Big Data Mining and Analytics · doi:10.26599/bdma.2025.9020034 · Cited by 12
CtxPipe: Context-aware Data Preparation Pipeline Construction for Machine Learning
Gao, H., Cai, S., Dinh, T. T. A., Huang, Z., & Ooi, B. C. (2024) · Proceedings of the ACM on Management of Data · doi:10.1145/3698831 · Cited by 19
“This process is optimized using different techniques, including genetic programming [42, 49], Bayesian optimization [16, 54], pruned tree search [30, 37], boosting [28], prediction uncertainty reduction [26], Reinforcement Learning (RL) [8, 11, 21], and some combination thereof [46].”
InfluentialMetaPrep: Data preparation pipelines recommendation via meta-learning
Zagatti, F. R., Silva, L. C., Silva, L. N. D. S., Sette, B. S., Caseli, H. D. M., Lucrédio, D., & Silva, D. F. (2021) · International Conference on Machine Learning and Applications · doi:10.1109/icmla52953.2021.00194 · Cited by 8
Unveiling Practices of Customer Service Content Curators of Conversational Agents
Candello, H. C. D. S. P., Pinhanez, C. S., Muller, M. J., & Wessel, M. S. (2022) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3555768 · Cited by 11
“Curation has also been studied in automated methods for selecting data in specific machine learning projects [10, 12, 93, 95].”
MethodAutoPrep: Natural Language Question-Aware Data Preparation with a Multi-Agent Framework
Fan, M., Fan, J., Tang, N., Cao, L., Li, G., & Du, X. (2025) · Proceedings of the VLDB Endowment · doi:10.14778/3748191.3748211 · Cited by 33
CatDB: Data-Catalog-Guided, LLM-Based Generation of Data-Centric ML Pipelines
Fathollahzadeh, S., Mansour, E., & Böehm, M. (2025) · Proceedings of the VLDB Endowment · doi:10.14778/3742728.3742754 · Cited by 18
“Tools like ActiveClean [44], CPClean [41], and Learn2Clean [15] focus on specific models or error types, such as missing values.”
Demystifying Artificial Intelligence for Data Preparation
Chai, C., Tang, N., Fan, J., & Luo, Y. (2023) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3555041.3589406 · Cited by 31
Detecting Data Accuracy Issues in Textual Geographical Data by a Clustering-based Approach
Pellegrino, M. A., Postiglione, L., & Scarano, V. (2020) · Joint International Conference on Data Science & Management of Data (CODS-COMAD) · doi:10.1145/3430984.3431031 · Cited by 7
“Machine learning techniques are becoming the core part of reactive data quality monitoring systems as they may provide fast and reliable data quality assurance [6] while reducing the human effort [9].”
BackgroundBrazdil, P., van Rijn, J. N., Soares, C., & Vanschoren, J. (2022) · Cognitive technologies · doi:10.1007/978-3-030-67024-5_14 · Cited by 5
A Data-Centric Approach for Reducing Carbon Emissions in Deep Learning
Anselmo, M., & Vitali, M. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-34560-9_8 · Cited by 10
Chen, S., He, Y., Cui, W., Fan, J., Ge, S., Zhang, H., Zhang, D., & Chaudhuri, S. (2024) · Proceedings of the ACM on Management of Data · doi:10.1145/3654925 · Cited by 24
ReClean: Reinforcement Learning for Automated Data Cleaning in ML Pipelines*
Abdelaal, M., Yayak, A. B., Klede, K., & Schöning, H. (2024) · IEEE International Conference on Data Engineering Workshops (ICDEW) · doi:10.1109/icdew61823.2024.00048 · Cited by 10
“To the best of our knowledge, ReClean is the first method that effectively leverages RL agents to select and combine data repair tools, to enhance downstream ML task performance.”
Active Reinforcement Learning for Data Preparation: Learn2Clean with Human-In-The-Loop
Berti‐Équille, L. (2020) · HAL · Cited by 3
Machine Learning Algorithms for Customer Churn Prediction in Telecoms Sectors
Migdady, A., Alnatsheh, A., & AlZoubi, O. (2025) · International Conference on Information and Communication Systems (ICICS) · doi:10.1109/icics65354.2025.11072960 · Cited by 3
“Berti-Equille [15],introduced Learn2Clean, a technology that offers a new method for preparing data.”
Analytics on Non-Normalized Data Sources: More Learning, Rather Than More Cleaning
Cvetkov-Iliev, A., Allauzen, A., & Varoquaux, G. P. (2022) · IEEE Access · doi:10.1109/access.2022.3168013 · Cited by 8
“First the analytic task provides super-vision [20]–[22], while cleaning needs examples of curated data.”
BackgroundExplanations for Machine Learning Pipelines under Data Drift
Hasan, J., & Pradhan, R. (2025) · Workshop on Human-In-the-Loop Data Analytics (HILDA) · doi:10.1145/3736733.3736744 · Cited by 3
“Moving forward, establishing causality between vulnerability and model degradation, supporting multi-drift analysis, and offering actionable, user-centered explanations remain essential to prioritizing robustness in ML deployment.”
Siddiqi, S., Phani, A., Kern, R., & Böehm, M. (2026) · ACM Transactions on Database Systems · doi:10.1145/3771766 · Cited by 5
“— Learn2Clean [9]: The recent Learn2Clean uses reinforcement learning (Q-learning) to find the best cleaning strategy for a configurable target ML application, without any other hyper-parameters or extension hooks.”
Mitigating Matching Scarcity in Recruitment Recommendation Domains
Cardoso, A., Mourão, F., & da Rocha, L. C. D. (2020) · Brazilian Symposium on Multimedia and the Web · doi:10.1145/3428658.3430968 · Cited by 2
“Dessa forma, a informação pode ser aplicada posteriormente em estratégias de recomendação tradicionais de maneira mais eficiente [2, 8].”
Background1st International Workshop on Data Assessment and Readiness for AI
Bandyopadhyay, B., Bandyopadhyay, S., Bedathur, S., Gupta, N., Mehta, S., Mujumdar, S., Parthasarathy, S., & Patel, H. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-75015-2_12 · Cited by 4
Leveraging Machine Learning for Effective Data Management
Sellami, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-662-70140-9_1 · Cited by 1
GA-PRE: A Genetic Algorithm-Based Automatic Data Preprocessing Algorithm
Jiao, J., & Yuan, L. (2025) · Genetic and Evolutionary Computation Conference (GECCO) · doi:10.1145/3712256.3726312 · Cited by 1
“Manual[18, 19] Not automated None Analytical[6, 8, 11] Not automated Moderate Deep Learning[2, 12, 20, 21] Semi-automated High AutoML[1, 4, 14, 15] Automated Moderate 2.1.1 Manual Methods.”
InfluentialRAC-GAN: Iterative Dual-Objective Over and Under Sampling for Imbalanced Datasets
Gupta, N., Mujumdar, S., Mittal, R. S., & Patel, H. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889166 · Cited by 1
Data quality checking for machine learning with MeSQuaL [demonstration paper]
Comignani, U., Novelli, N., & Berti‐Équille, L. (2020) · Cited by 1
Modeling Adaptive Data Analysis Pipelines for Crowd-Enhanced Processes
Cappiello, C., Pernici, B., & Vitali, M. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-89022-3_3
Explainable Error Detection Method for Structured Data using HoloDetect framework
Khorasani, A. M., Ghassabi, S., Behkamal, B., & Milani, M. (2023) · International Conference on Computer and Knowledge Engineering · doi:10.1109/iccke60553.2023.10326276
“The first framework is CleanEx [5] which adds an extent ion to the Learn2Clean [2] model for explanation.”
MethodBiliri, E., Lampathaki, F., Mandilaras, G., Prieto-Roig, A., Calabresi, M., Branco, R., & Gkolemis, V. (2023) · IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC) · doi:10.1109/ice/itmc58018.2023.10332317 · Cited by 4
LucidScript: Bottom-Up Standardization for Data Preparation
Lai, E. Y., Lou, Y., Youngmann, B., & Cafarella, M. (2024) · Proceedings of the VLDB Endowment · doi:10.14778/3685800.3685864
Lässig, N., & Herschel, M. (2025) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde65448.2025.00096
“We acknowledge that the time-consuming data preparation optimization could be improved using dedicated frameworks optimizing the preparation process (e.g., [94]–[96]).”
SwiftDP: An Efficient Framework for Automated Data Preparation Pipeline Generation
Li, L., Zhang, Y., & Wang, N. (2025) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde65448.2025.00355
TORepair: Diffusion-Based Task-Oriented Error Repair Via Differentiable Bi-Level Optimization
Ni, W., Miao, X., Zhao, X., Wu, Y., & Yin, J. (2026) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde65706.2026.00198
“Another line of research explores automated ML data pipelines [24], [42], [44], [64]–[66] primarily focus on macro-level pipeline configurations,which orchestrate various operations like cleaning and transformation to maximize down-stream performance.”
Li, M., Bu, F., Zhang, Y., & Sun, F. (2026) · Sustainability · doi:10.3390/su18189277
On the impact of low-quality activity labels on data-driven process analytics
Comuzzi, M., Kim, S., Ko, J., Cappiello, C., Salamov, M., & Pernici, B. (2026) · International Journal of Data Science and Analytics · doi:10.1007/s41060-026-01334-z
“In [16, 23], the optimal data preparation pipeline design is based on a reinforcement learning approach.”
Evidence-based data preparation pipeline design for data-centric AI
Sancricca, C., Sabella, M., & Cappiello, C. (2026) · Future Generation Computer Systems · doi:10.1016/j.future.2026.108893
Metadata
- Authors
- Berti-Equille, L.
- Affiliations
- Institute of Research for Development (IRD), France, France
- Venue
- The World Wide Web Conference
- Publisher
- ACM
- Published
- 2019-05-13
- Pages
- 2580-2586
- OpenAlex
- W2914436728
- Cited by
- 54
- References
- 21
- Citations by year
- 2026: 4 · 2025: 12 · 2024: 8 · 2023: 10 · 2022: 6 · 2021: 6 · 2020: 6 · 2019: 2
- FWCI
- 7.16
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- International World Wide Web Conference (WWW) (publishedVersion, cc-by)
- HAL AMU (submittedVersion)
- Funders
- Agence Nationale de la Recherche (ANR-18-CE23-0002)
- Topics
- Physical Sciences › Computer Science › Information Systems › Data Quality and Management
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Physical Sciences › Computer Science › Information Systems › Web Data Mining and Analysis
- Keywords
- data cleaning, task sequencing, data preprocessing, automated machine learning, data curation, data quality, web data
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.