Skip to content
No description, website, or topics provided.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Raha: A Configuration-Free Error Detection System

Detecting erroneous values is a key step in data cleaning. Error detection algorithms usually require a user to provide input configurations in the form of integrity constraints or statistical parameters. However, providing a complete, yet correct, set of configurations for each new dataset is tedious and error-prone, as the user has to know about both the dataset and the error detection algorithms upfront.

Raha is a new configuration-free error detection system. The basic idea is that by generating a reasonably limited number of configurations for a set of error detection algorithms covering different types of data errors, we can generate an expressive feature vector for each entry in the dataset. Leveraging these feature vectors, we propose a novel sampling and classification approach that effectively chooses the most representative values for training. Furthermore, Raha can exploit historical data, if available, to filter out irrelevant error detection algorithms and configurations.


This project is implemented on top of abstraction layer.



This folder contains input datasets.


This folder contains the underlying data cleaning tools.


This folder contains the outputted results.

This file contains the implementation of the dataset class.

This file contains the implementation of the data cleaning tool class.

This file contains the implementation of the main application.


This file demonstrates the system in a Jupyter notebook.


You can find more information about this project and the authors here.

You can also use the following bib entry to cite this project/paper.

  title={Raha: A configuration-free error detection system},
  author={Mahdavi, Mohammad and Abedjan, Ziawasch and Castro Fernandez, Raul and Madden, Samuel and Ouzzani, Mourad and Stonebraker, Michael and Tang, Nan},
You can’t perform that action at this time.