Statcheck

Advancing ‘statcheck’ by using Natural Language Processing

In close collaboration with colleague Willem Sleegers, Michèle Nuijten developed statcheck, a ‘spellchecker’ for statistics. With this open-access tool researchers extract the statistics reported in a paper and automatically flag inconsistencies. However, statcheck cannot catch all errors, because it detects the reported statistics in a specific way.

This project aims to improve statcheck’s searching algorithm to make it ‘smarter’ in recognizing statistical results so that it can also spot errors in other scientific fields. Moreover, the project team wants to expand statcheck to improve scientific quality on an even larger scale.

By improving statcheck’s sensitivity in identifying statistical results, the Netherlands eScience Center expects to make a valuable contribution to open science.

Participating organisations

Tilburg University
Netherlands eScience Center
Social Sciences & Humanities
Social Sciences & Humanities

Team

MN
Michèle Nuijten
WS
Willem Sleegers
Rena Bakhshi
Programme manager
Netherlands eScience Center
Jisk Attema
Programme Manager
Netherlands eScience Center
Dafne van Kuppevelt
Dafne van Kuppevelt
Consultant
Netherlands eScience Center

Related projects

TMSR

Text Mining Systematic Review

Updated 7 months ago
In progress