KATARA: A Data Cleaning System Powered by Knowledge Bases and Crowdsourcing

By Xu Chu, J. Morcos, Ihab F. Ilyas, Mourad Ouzzani, Paolo Papotti, Nan Tang, Yin Ye. 2015

319 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Information Systems, Management Science and Operations Research, Knowledge Graphs & Representation, ML Systems & Hardware

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Information Systems

In Management Science and Operations Research

In Knowledge Graphs & Representation

In ML Systems & Hardware

Citations in the corpus

AutoML: A Survey of the State-of-the-Art (2019, background): “As an alternative, some studies propose possible automated data-cleaning processes [78, 79, 80, 81].”
Benchmark and Survey of Automated Machine Learning Frameworks (2019, background): “Using different representations of expert knowledge, like integrity constraints or first order logic, low quality data can be detected and corrected automatically (Raman & Hellerstein, 2001; Hellerstein, 2008; Chu et al., 2015, 2016).”
Survey on Automated Machine Learning (2019, background): “Using different representations of expert knowledge, like integrity constraints or first order logic, low quality data can be automatically detected and corrected (Raman and Hellerstein, 2001; Hellerstein, 2008; Chu et al., 2015, 2016).”
BoostClean: Automated Error Detection and Repair for Machine Learning (2017, background): “Several tools have been proposed to reduce human burden including automatically generating explanations [50], building more robust interfaces [3,26], and more sophisticated human-in-the-loop processing with crowds [15,21,22,36,40,52].”
HoloClean: Holistic Data Repairs with Probabilistic Inference (2017, methodology): “• KATARA [13]: This is a knowledge base (KB) powered data cleaning system that, given a dataset and a KB, interprets table semantics to align it with the KB, identiﬁes correct and incorrect data, and generates repairs for incorrect data.”
Towards Automated Data Cleaning Workflows (2019, methodology): “There has already been a large body of work on data cleaning [4, 6, 23, 5].”
Can Foundation Models Wrangle Your Data? (2022, methodology): “These are usually hard-coded with human-engineered rules or external knowledge bases [24, 85].”
Trends in Cleaning Relational Data: Consistency and Deduplication (2015, methodology): “KATARA [28] aims at producing accurate repairs by relying on two authoritative data sources, namely, knowledge bases (kbs), and domain experts.”
SampleClean: Fast and Reliable Analytics on Dirty Data (2015, background): “While crowdsourcing is an increasingly viable option for correcting some types of errors [29, 22, 48, 24, 38, 4, 14], it comes at the signiﬁcant cost of additional latency and the overhead of managing human workers.”
Cleaning Crowdsourced Labels Using Oracles For Statistical Classification (2018, background): “One category is to leverage humans (either crowd workers or experts) to solve a particular data-cleaning problem, such as entity resolution [9,14,41,43], missing value imputation [30], and data repairing [8,44].”
Detecting Data Errors: Where are we and what needs to be done? (2016, methodology): “Finally, while OpenRefine and Trifacta focus on syntactic patterns, Katara focuses on semantic patterns matched against a knowledge base.”