======================== EVALD 2.0 for Foreigners ======================== EVALD 2.0 for Foreigners is a software for automatic evaluation of surface coherence (cohesion) in Czech texts written by non-native speakers of Czech. It is an update of the previous version, EVALD 1.0 for Foreigners. The evaluation part (the server) is implemented in Treex (http://ufal.cz/treex), a highly modular NLP framework written in the Perl programming language, and uses the Weka toolkit (http://www.cs.waikato.ac.nz/ml/weka/) for the final prediction of a coherence mark. It can be used directly from a command line or as a backend server for a client. The client part is implemented as a web server, accessible with a web browser. ================================================================ There are three possible ways of using EVALD 2.0 for Foreigners: ================================================================ 1. interactively as a web demo and RESTful web service hosted at the LINDAT/CLARIN server, 2. interactively but locally, with both the server and the client running on the same machine (or two machines in the same network), 3. in a batch mode run on the local machine. ================================================= Ad 1) No installation is needed in this case; in a web browser (such as Firefox or Chrome), go to https://lindat.mff.cuni.cz/services/evald-foreign/. ================================================= Ad 2) Both the backend server and the client components are distributed via the Docker software (https://www.docker.com/), which needs to be installed first. Docker greatly simplifies the installation process of the two components and allows them to be run on Linux-based operating systems, Windows 10, as well as Mac OS X. To download both the server and the client, run the following commands: docker pull ufal/evald.treex-server:2.0 docker pull ufal/evald.php-server:2.0 See the EVALD 2.0 for Foreigners project web pages (http://ufal.mff.cuni.cz/evald/documentation) for further instructions. ================================================= Ad 3) The dockerized backend component can be run locally also in a batch mode. In a command line, the user must specify paths to input and output directories. All files in the input directory are then processed and the result is stored in the output directory. In order (for more experienced users) to gain better control over the processing pipeline, the backend application can be installed also without being wrapped by a Docker container. Treex (http://ufal.cz/treex) needs to be installed on the local machine, along with all (mostly CPAN) dependencies for the Czech text analysis (please note that in this case, some parts of the Czech analysis can only be run on Linux-based systems). Treex must be in the revision tagged as EVALD_2.0 (https://github.com/ufal/treex/releases/tag/EVALD_2.0). In addition, Vowpal Wabbit 8.1.1 (https://github.com/JohnLangford/vowpal_wabbit/releases/tag/8.1.1) must be installed to the location installed_tools/ml/vowpal_wabbit-v8.1-3cf3f692/ relative to the Treex Share directory. The Treex scenario to be run is a part of this Lindat EVALD 2.0 for Foreigners distribution (file Evald-2.0-Foreign.scen). See the EVALD 2.0 for Foreigners project web pages (http://ufal.mff.cuni.cz/evald/documentation) for further instructions. ============ Further Info ============ EVALD 2.0 for Foreigners project web pages: https://ufal.mff.cuni.cz/evald If you have questions or need technical support, please contact evald@ufal.mff.cuni.cz. ==================================== How to Cite EVALD 2.0 for Foreigners ==================================== Novák Michal, Rysová Kateřina, Mírovský Jiří, Rysová Magdaléna, Hajičová Eva: EVALD 2.0 for Foreigners. Data/software, LINDAT/CLARIN digital library, Prague, Czech Republic, http://hdl.handle.net/11234/1-2509, 2017. There are also papers describing the related research and experiments: Novák Michal, Rysová Kateřina, Rysová Magdaléna, Mírovský Jiří: Incorporating Coreference to Automatic Evaluation of Coherence in Essays. In: Statistical Language and Speech Processing. Cham, Switzerland: Springer International Publishing. ISBN 978-3-319-68455-0, ISSN 1611-3349, pp. 58–69, 2017. Rysová Kateřina, Rysová Magdaléna, Mírovský Jiří, Novák Michal: Introducing EVALD – Software Applications for Automatic Evaluation of Discourse in Czech. In: International Conference Recent Advances in Natural Language Processing 2017. Šumen, Bulgaria. ISSN 1313-8502, pp. 634–641, 2017. =============== Acknowledgement =============== EVALD 2.0 for Foreigners was developed at the Institute of Formal and Applied Linguistics (ÚFAL, http://ufal.mff.cuni.cz/), Faculty of Mathematics and Physics, Charles University, with the financial support of the Ministry of Culture of the Czech Republic, project Automatic Evaluation of Text Coherence in Czech (DG16P02B016, http://ufal.mff.cuni.cz/grants/evald-evaluator-discourse).