E:Calm Resource: a Resource for Studying Texts Produced by French Pupils and Students

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 4327–4332 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 4327 ´E:CALM Resource: a Resource for Studying Texts Produced by French Pupils and Students Lydia-Mai Ho-Dac, Serge Fleury, Claude Ponton CLLE – CNRS – University Toulouse Jean Jaur`es CLESTHIA – University of Paris 3 Sorbonne Nouvelle LIDILEM – University of Grenoble Alpes lydia-mai.ho-dac@univ-tlse2.fr, serge.fleury@sorbonne-nouvelle.fr, claude.ponton@univ-grenoble-alpes.fr Abstract The ´E:CALM resource is constructed from French student texts produced in a variety of usual contexts of teaching. The distinction of the ´E:CALM resource is to provide an ecological data set that gives a broad overview of texts written at elementary school, high school and university. This paper describes the whole data processing: encoding of the main graphical aspects of the handwritten primary sources according to the TEI-P5 norm; spelling standardizing; POS tagging and syntactic parsing evaluation. Keywords: education, handwritten encoding, spelling 1. Introduction The ´E:CALM resource is constructed from French student texts produced in a variety of usual contexts of teaching. The key feature of the ´E:CALM resource is to provide an ecological digital data set that gives a broad overview of texts written at elementary school, high school and univer- sity (Doquet et al., 2017b). The advantages of such a resource are multidisciplinary. From a scientific point of view, the ´E:CALM resource pro- vides a valuable data set for the digital humanities and writ- ing sciences, since it allows the unparalleled possibility to observe the acquisition of literacy and especially writing skills through all the education levels. From an educa- tion point of view, it could be used for teaching literacy by working with students on real-life texts and by focusing on attested misspellings and coherence issues. In addition, it could be used for identifying the main problems and doubts encountered by students at each grade. As for NLP and corpus linguistics, such a resource constitutes a good ex- perimental field for evaluating and adapting models, meth- ods and tools on the (manual or automatic) annotation of non-standard corpora. The decision to encode handwritten student texts according to the TEI-P5 norm ensures con- sistency, practicability, compatibility with a broad range of corpus tools and data exchange facilities (Burnard, 2007). Figure 1 gives an example of a primary source that com- posed the ´E:CALM resource. Figure 1: Example of primary source collected from 4th grade pupils. The main part of the resource is made up with such hand- written school works. As the figure shows, the collected manuscripts may contain student revisions (e.g. erasures) and teacher comments (here in red). This paper describes the whole data processing established for digitizing and standardizing such manuscripts in order to provide a new resource for NLP, Corpus Linguistics and Education. 2. Related Works The Lancaster Corpus of Children’s Project Writing (Smith et al., 1998) is one of the first children’s corpora transcribed and available online. It is the first in the field of school cor- pora, bringing together a …

Content truncated to 500 words