/numerical_outlier_detection

This code in this repository implements the paper *Detecting Incorrect Numerical Data in DBpedia* by Dominik Wienand and Heiko Paulheim.

Primary LanguagePython

numerical_outlier_detection

This code in this repository implements the paper Detecting Incorrect Numerical Data in DBpedia by Dominik Wienand and Heiko Paulheim.

##Requirements To install the requirements

pip3 install -r requirements.txt

Execution

To run the code,

python3 main.py

Data

The data directory has the outliers stored for the queries which have been run.

The data directory has sub-directories, where each folder is for a query which has been run. The country sub-directory has the outliers calculated using different methods in different files. There is a file (country.parsing_exception) which has the list of the countries, whose populations are not an integer (we expect the population to be an integer).