Joke van de Plassche’s research while affiliated with University of Amsterdam and other places

This paper describes a method of analysing words through morphological decomposition when the lexicon is incomplete. The method is used within a text-to-speech system to help generate pronunciations of unknown words. The method is achieved within a general morphological analyser system using Koskenniemi twolevel rules. Keywords: Morphology, incomplete lexicon, text-to-speech systems Background When a text-to-speech synthesis system is used, it is likely that the text being processed will contain a few words which do not appear in the lexicon as entries in their own right. If the lexicon consists only of whole-word entries, then the method for producing a pronunciation for such "unknown" words is simply to pass them through a set of letter-to-sound rules followed by word stress assignment rules and vowel reduction rules. The resulting pronunciation may well be inaccurate, particularly in English (which often shows a poor relationship between spelling and pronunciation). In addition, ...

Formal specification languages can improve the development of knowledge-based systems, but several problems limit their usefulness. (ML)², a formal language based on the knowledge models used in the CommonKADS KBS development method, avoids many of these problems. We describe a study that investigated the usability of a formal KBS modeling language, (ML)². To analyze (ML)² usability, we first designed a set of evaluation criteria. Then we performed a small case study, constructing an expertise model in (ML) ², to test and refine these criteria. Subsequently, we used (ML)² to construct a second model, which formed the basis for our language evaluation. Altogether, we performed three case studies, which we used to score (ML)² with our evaluation criteria

Describes Spengels, a knowledge-based tutoring system for the conjugation and spelling of English verbs for Dutch writers. The student has to fill in gapped-sentences. If an incorrect answer is entered, the system locates the point where the student deviated from the path through a spelling algorithm that leads to the correct answer. The student is either guided from that particular point through the spelling algorithm or given hints about what went wrong. The techniques that are applied for diagnosis of knowledge and performance errors are model driven. An enumerative and a reconstructive theory of bugs are combined to diagnose errors and to build a student model. Continuous knowledge diagnosis enables Spengels to present only those problems which are a challenge to the student. At the end of the session, a short overview of the student's performance is provided. (PsycINFO Database Record (c) 2012 APA, all rights reserved)