Automating Multi-Level Annotations of Orthographic Properties of German Words and Children’s Spelling Errors

Abstract

This paper presents the automatic annotation of orthographic properties of German words and spelling errors in texts of German primary school children according to a new multi-layered annotation scheme [1]. The scheme is closely linked to the principles of the German writing system and is supposed to allow the pursuit of new research questions concerning the relationship between spelling errors of competent and less competent spellers and the regularities of the German graphematic system. A novelty of the automatic annotation is that it takes an intended, correctly spelled word as input and applies a set of rules to generate a list of error candidates containing systematic spelling errors. As a further novelty, the annotation of additional word-and error-related properties is presented such as whether the spelling error changes the word's pronunciation and whether a spelling can be derived from a related word form. This gives rise to more detailed analyses of the errors but also allows us to develop an application for learners that generates automatic advice for the correct spelling. A first evaluation shows that the automatic annotation of the presented categories and features can come close to human annotations.

Generating Feedback for English Foreign Language Exercises

“…However, manually specifying such information quickly becomes infeasible even for relatively closed task types, as shown by Nagata (2009, p. 563) in the context of the Robo-Sensei system. Laarmann-Quante (2016) proposes an approach for the diagnosis of spelling errors in the writing of German children that was independently developed but is conceptually similar to the perspective we pursue in this paper. Instead of attempting to process the erroneous forms directly, LaarmannQuante obtains phonological analyses for correct spellings and uses rewrite rules that emulate typical misspellings to derive alternatives that can then be matched against actual input.…”

Towards a Tool for Automatic Spelling Error Analysis and Feedback Generation for Freely Written German Texts Produced by Primary School Children

“…The automatic annotation of a word's properties according to our annotation scheme described in the previous section is carried out with the help of the web service G2P of the Bavarian Archive of Speech Signals (BAS) 7 [19,20] and our own processes. These are described in [9] but to convey how the spelling analysis of our proposed application works, the basic procedures will be sketched here again.…”

“…To obtain these character units which we call phoneme-corresponding units or PCUs 8 , we firstly use string alignment via weighted Levenshtein distance. 9 The resulting 1:1 (or 1:0, 0:1) correspondences between characters and phonemes are then rearranged via rules to get the correct PCUs.…”

“…Furthermore, it is stored in which PCU the error occurred. If there are multiple errors and/or unsystematic errors in a word, all combinations of all error candidates of a word are computed and compared with the original spelling, see [9] for details.…”

“…Our goal is to facilitate a systematic analysis of all the errors in a text and to take different features beyond a one-dimensional error categorization into account. For instance, we want the application to automatically provide feedback about whether the pronunciation of the word changes with a misspelling or how one could otherwise arrive at the correct spelling using a previously developed annotation scheme for spelling errors and its automatic application [8,9]. The aim of this paper is to show how these rather abstract annotations can be presented in a user-friendly way so that they could be used e.g.…”