gilzoide / coh-metrix-dementia

Coh-Metrix-Port is an adaptation of the Coh-Metrix text analysis tool to the Brazilian Portuguese language.

Coh-Metrix-Port 2.0

Coh-Metrix-Port is an adaptation of the Coh-Metrix text analysis tool to the Brazilian Portuguese language. Coh-Metrix (http://cohmetrix.memphis.edu/cohmetrixpr/index.html), originally developed at the University of Memphis, is a Web-based tool that calculates textual metrics intended to evaluate cohesion, coherence and readability of texts. It replaces and extends traditional, surface-level readability formulas by using a variety of NLP tools and resources to more accurately evaluate text coherence.

Coh-Metrix-Port had already been implemented in Ruby, and used in several research scenarios. For more details, see Scarton & Aluísio [2010] and Scarton et al. [2010]. This version is called Coh-Metrix-Port 1.0, and its code has not been released public.

The 2.0 version is a complete, from-scratch rewrite of Coh-Metrix-Port 1.0 in Python. It uses updated resources and tools for improved performance, and includes metrics previously unavailable in Coh-Metrix-Port 1.0.

References

Scarton, C. & Aluísio, S. [2010]. Análise da Inteligibilidade de textos via ferramentas de Processamento de Língua Natural: adaptando as métricas do Coh-Metrix para o Português. Linguamática, 2(1), 45–62.

Scarton, C., Gasperin, C., & Aluisio, S. (2010). Revisiting the Readability Assessment of Texts in Portuguese. In A. Kuri-Morales & G. Simari (Eds.), Advances in Artificial Intelligence – IBERAMIA 2010, volume 6433 of Lecture Notes in Computer Science (pp. 306–315). Springer Berlin Heidelberg.

Dependencies

For Coh-Metrix-Port to run properly, you must first install the Python dependencies (optionally in a virtualenv):

$ pip install -r requirements.txt

License

This software is released under the GNU General Public License, version 3. For the complete text of this license, see file LICENSE.md.