Bibliography
Articles
Ahmed, Mehreen: A Note on Phrase Structure Analysis and Design Implication for ICALL. In: Computer Assisted Language Learning. 15 (4), 2002, pp 423-433
Note: Exploiting Chomsky's Universal Grammar this research examines the effect of phrase structure rules on the development of grammatical knowledge in Bangladeshi learners of English as a second language. The masking technique of McWhinney motivates the use of the link-grammar syntactic parser of Sleator and Temperley used in the study. The results indicate that the learners' parameter were reset from L1 to L2 as a consequence of the overt presentation of phrase structure rules. The implication of the study for the design of intelligent computer-assisted language learning systems is discussed. The results however are drawn on rather weak evidence as only observations, interviews and open-ended questionaires were used.
Carbonell, Jaime R.: AI in CAI: An Artificial Intelligence Approach to Computer-Assisted Instruction. In: IEEE Transactions on Man-Machine Systems. 11 (4), 1970, pp 190-202
Note: This article is one of the very early texts about the application of methods of Computational Linguistics and Artificial Intelligence to CALL. It discusses the AI techniques from that time which make it possible to incorporate meaningful tutorial dialogues in computer assisted instruction programs. It also describes the SCHOLAR program, which incorporates an information network of facts, concepts and procedures, which enables it to carry on a mixed initiative dialogue between student and computer, handling questions and answers from both sides. The content is geographical issues in South America.
Eidecker, Martina Elisabeth: Anwendungsweisen multimedialer Computerprogramme im Fremdsprachennterrricht. In: Deutsch als Fremdsprache. 2, 1996, pp 109–115
Note: CyberBuch ist eine multimediale Software, die Studenten den Zugang zu authentischen Lesetexten erleichtern soll. Sie unterstützt die Entwicklung von Lesestrategien, indem sie nicht nur den Text anbietet, sondern auch „Organisatoren", die den Text kontextualisieren (Filme, Fotografien und Hintergrundinformationen). In Übereinstimmung mit dem Natürlichen Ansatz in der Spracherwerbstheorie finden Sprachaneignung und Spracherlernen nur in der Zielsprache statt.
Über die Vorstellung dieses Programmpakets hinaus wird aber auch kaum ein neuer Aspekt von Computer und FU eingebracht. Aus heutiger Sicht ist der Artikel damit zum großen Teil veraltet, da es nun viele CyberBuch ähnliche System gibt.
Gamper, Johann ; Knapp, Judith: A Review of Intelligent CALL Systems. In: Computer Assisted Language Learning. 15 (4), 2002, pp 329-342
Note: This paper provides an overview about intelligent computer-assisted language learning. The most advanced systems have been investigated and classified along five dimensions: supported languages, AI techniques, language skills, language elements and availability. The paper concludes with a discussion about outstanding problems. All the important systems after 1995 are included and a good overview is given in a comprehensive table. Unfortunately only one major publication per project is mentioned which limits the the bibliography a bit.
Granger, Sylviane ; Vandeventer, Anne ; Hamel, Marie-Josee: Analyse de corpus d'apprenants pour l'ELAO base sur le TAL. In: T.A.L.. 42(2), 2001, pp 609-621
Note: In this article Granger et al. report about the FRIDA corpus of learner's French and the combination with the FreeText project in which a grammar checker was developed. First the construction and annotation of the corpus is described and in the second part an evaluation of the parser used in the FreeText project is presented. Finally some conclusions with regard to the usage in language learning and didactic issues are discussed. Since a lot of topics need to covered there are not too many details in the text. But it gives a good impression of the chances of using certain types of CL technology for CALL.
Heift, Trude: Error-Specific and Individualized Feedback in a Web-based Language Tutoring System: Do They Read It?. In: ReCALL. 13 (2), 2001, pp 129-142
Note: In this article Heift presents a user study with 33 students using her system E-Tutor. Even though the exercises are fairly restricted the systems uses a HPSG-based parser for the generation of "meta-linguistic" feedback. The learners were able to decide whether they wanted to read specific feedback or leave it at general remarks. The study shows that the student actually read the more specific texts and even paid increasing attention to the feedback with more iterations of the exercises. Even though no "post-test" was conducted the study shows that students seem to pay attention to the parser generated feedback.
L'haire, Sebastien ; Vandeventer Faltin, Anne: Error Diagnosis in the FreeText Project. In: CALICO Journal. 20(3), 2003, pp 481-495
Note: In this article the authors present the results from the FreeText project. The system developed has a different approach of error diagnosis on every linguistic level which are presented level by level (orthography, syntax and semantics). Based on an already existing large coverage parser the syntactic module is able to identify quite a number of errors. Note also, that here one of the few approaches to tackling semantic errors is presented.
Ludewig, Petra: LogoTax - un outil exploratoire pour l'etude de collocations en corpus. In: T.A.L.. 42(2), 2001, pp 623-642
Note: In this paper the LogoTax system is presented. LogoTax is a kind of personal electronic dictionary which supports the learner in entering items into a personal database. To reach this goal corpora are analysed syntactically in order to provide feedback especially for collocations. The learner should be able to identify the usage patterns of a collocation through the corpus examples which additionally are presented after a syntactical evaluation. This process is described indepth. Note however that the system can also store any other kind of entry the user whishes to keep.
Reuer, Veit: Error Recognition and Feedback with Lexical Functional Grammar. In: CALICO Journal. 20(3), 2003, pp 497-512
Note:
This paper describes the error recognition module of an interactive ICALL system with a special focus on the underlying grammar theory. It is argued that the theory of Lexical Functional Grammar (LFG) is well suited to be used in the parsing and error recognition module of the system as well as to provide intelligent feedback to learners. The concepts and structures used in LFG closely resemble the descriptive knowledge of language learners about a language, and, therefore, the results of an automatic analysis can easily be translated from a computationally tractable form to language easily understood by the learner. Note however that the formal aspects of the parsing algorithm are left out sometimes in order to highlight the possibilities for the actual learner.
Vandeventer, Anne: Creating a grammar checker for CALL by constraint relaxation: a feasibility study. In: ReCALL. 13(1), 2001, pp 110-120
Note: This study in the context of the FreeText project explores the usability of the Fips parser for error recognition. The paper reports on the recognition of agreement errors in French sentences taken from the FRIDA corpus also mentioned in Granger et al. 2001. The specific technique used is a kind of constraint relaxation. The relevant feature receives a marking that an error has been encountered. In total a precision above 50 per cent and a recall above 70 per cent was reached. However a large number of false positives were encountered which is discussed in detail. Note that the technical details are only sketched.
Weischedel, Ralph M. ; Sondheimer, Norman K.: Meta-rules as a Basis for Processing Ill-Formed Input. In: Computational Linguistics. 9, 1983, pp 161-177
http://acl.ldc.upenn.edu/J/J83/J83-3003.pdfNote: (checked 6.2004) In this paper meta-rules and a control structure under which they are invoked as a framework for processing ill-formed input are proposed. The LHS refers to an error encountered during normal parsing and the RHS relaxes the violation. Examples discussed in the paper include violated grammatical tests, omitted articles, homonyms, spelling/typographical errors, unknown words, violated selection restrictions, personification, and metonymy. Also the important distinction between "absolutely illformed" and "relatively illformed" is made. Note that the parser is based on an ATN which seems not so common nowadays.
Wolff, Dieter: Neue Technologien und fremdsprachliches Lernen - Versuch einer Bestandsaufnahme. In: DaF. 35, 1998, pp 136-140 + 205-211
Note: In diesem zweiteiligen Artikel stellt Wolff die Ideen vor, die später in Rüschoff und Wolff 1999 als Buch ausführlicher dargestellt wurden. Grundlage ist hier der Versuch einer Strukturierung des Bereichs CALL mit einem Fokus auf konstruktivistische Ansätze. Hervorzuheben ist hier der Versuch der Strukturierung von Computerprogrammen, die einerseits damit einer Bewertung unterzogen werden und andererseits dahingehend beurteilt werden, wie geeignet sie für konstruktivistische Ansichten sind.
AV
Handke, Jürgen ; Intemann, Frauke:
The Interactive Introduction to Linguistics. Ismaning : Hueber, 1999 Version 1.10
Note: CD-ROM for Win9x/NT. This CD-Rom which came out as Version 2.00 in 2000 presents a multi media based introduction to linguistics. Most of the core areas of linguistics from phonetics to semantics are covered on an introductory level. Therefore it is well suited for beginning students of linguistics. It uses the possibilities of multi media very well especially (as one would expect) in the area of phonetics and phonology. Also the chapter on the language of the world profits from the use of a CD-Rom in order to protray the different sounds of the languages. Note that this CD-Rom is recommended as a supplement to a introductory course.
Heringer, Hans Jürgen:
Aus Fehlern lernen. Augsburg : Universität Augsburg, 1995
http://www.philhist.uni-augsburg.de/faecher/germanis/daf/forschung/fehler/index.htmlNote: CD-ROM for Win9x/NT. Diese CD-Rom enthält ein einfaches Konzept zur Präsentation von Fehlern aus einem annotierten Fehlerkorpus für das Deutsche. Die Aufgabe besteht in der Bestimmung des Fehlers durch den Lerner. Anschließend kann dann nachgeschlagen werden, ob der Fehler richtig erkannt wurde. Interessant an dieser CD-Rom ist die Tatsache, dass die Sätze alle authentisch sind, allerdings so korrigiert worden sind, dass jeder Satz nur einen Fehler enthält. Diese Datenbank lässt sich auch nach XML konvertieren und so für eine automatische Analyse nutzen.
Books
Atwell, Eric:
The Language Machine. London : British Council, 1999
http://www.comp.leeds.ac.uk/eric/atwell99.pdfNote: (checked 5.2004)
In this booklet Atwell tries to foresee what the future will bring with regard to language technology and language learning. The leading question is whether language learning is still relevant if comprehensive translation systems have been developed in the future. As this book was written for the British Council it is written in a rather popular style. On the one hand the current state-of-the-art in speech and language technology is explained and on the other hand the possibilities for future developments are discussed. The booklet also addresses the problems of CL in popular terms. Finally it also highlights the industry and research scenarios with a specific focus on the UK.
Bloom, Benjamin (Ed.):
Taxonomy of Educational Objectives. The Classification of Educational Goals. Handbook I, Cognitive Domain. New York : Longman, 1956
Note: This is THE classical work with respect to educational goals. This is still the major reference and to my knowledge no really better solution has been found. Even though it almost half a century old the classification can still be used. Of course most other aspects seem outdated nowadays.
Fechner, Jürgen (Ed.):
Neue Wege im computergestützten Frendsprachunterricht. Berlin : Langenscheidt, 1994
Goerz, Günther ; Rollinger, Claus-Rainer ; Schneeberger, Josef (Ed.):
Handbuch der künstlichen Intelligenz. 3. München : Oldenbourg, 2000
Note: Diese Buch bietet die einzige mir bekannte deutschsprachige Übersicht über Künstliche Intelligenz. Es teilt sich in die folgenden Kapitel auf: Kognition, Neuronale Netze, Suche und Constraints, Wissensrepräsentation, Logik und automatisches Beweisen, Unsicheres und vages Wissen, Wissen über Raum und Zeit, Fallbasiertes Schließen und modellbasierte Systeme, Planen, Maschinelles Lernen und Data Mining, Sprachverarbeitung, Bildverstehen, Robotik und Software-Agenten. Die Themen werden aber sehr unterschiedlich angegangen. Manchmal sind Vorkenntnisse unabdingbar und manchmal werden auch grundlegende Aspekte ausführlich erläutert.
Greer, Jim (Ed.):
Student modelling: the key to individualized knowledge based instruction. Berlin : Springer, 1994
Grüner, Margit ; Hassert, Timm:
Computer im Deutschunterricht. München : Langenscheidt, 2000
Note: Dieses Buch beschreibt in sehr einfacher Form die Möglichkeiten der
Higgins, John:
Computers and English language learning. Oxford : Intellect, 1995
Holland, V. Melissa ; Kaplan, Jonathan D. ; Sams, Michelle R. (Ed.):
Intelligent Language Tutors. Mahwah, NJ : Erlbaum, 1995
Note: As opposed to Yazdani et al. (1992) this book concentrates on contributions to the development of ICALL systems from North America. The book is divided into 4 sections: Text-Based Language Tutors and Learning Environments, Dialogue-Based Language Games, Graphics-Based Language Tutors and Learning Environments, and Theoretical Issues in Language Tutor Design and Assesment. Notably the book contains descriptions of the ALLP project, the ALICE-chan system and the program 'Herr Kommissar'. There also a few articles from scientists who are not normally connected with NLP, such as one by Nina Garrett and one by Brian McWhinney.
Jager, Sake ; Nerbonne, John ; Van Essen, Arthur (Ed.):
Language teaching and language technology. Lisse : Swets and Zeitlinger, 1998
Note: This book grew out of a workshop held at the University of Groningen in April 1997 a contains selected papers from this workshop. The articles are divided into seven sections: Speaking (3 papers), vocabulary (3 papers), grammar (3 papers), reading, writing and testing (4 papers), distance learning (3 papers), users: models and studies (5 papers), and reflections and visions (2 papers). The length of the papers ranges from 3 pages to 15 pages. Some projects are presented in more than one paper, notably the Glosser project and the RECALL project with 3 papers each. What the book clearly demonstrates as opposed to e.g. Holland et al. (1995) is that here a some user studies are already included. I think this demonstrates the change from well designed systems but too slow to be used (ALLP) to systems like Glosser, which may contain less NLP technology but was actually used by language learners.
Jung, Udo O.H.:
An international bibliography of computer-assisted language learning with annotations in German. 1. Edition. Frankfurt am Main : Lang, 1988
Krüger-Thielmann, Karin:
Wissensbasierte Sprachlernsysteme. Tübingen : Narr, 1992
Note: Dieses ist die Dissertation von Krüger Thielmann, in der Ansätze zur Fehlererkennung mit Hilfe unterschiedlicher Parsingstrategien auf der Basis von PROLOG beschrieben werden. Gegenüber sehr umfassenden Parsern bechränkt sich dieser Ansatz auf die Beschreibung des Zahlensystems im Französischen.
Levy, Michael:
Computer-Assisted Language Learning - Context and Conceptualization. Oxford : Clarendon Press, 1997
Note: Michael Levy's book tries to interpret the development of CALL and also sheds some light on the history of ICALL. The historical development is then used to argue for new strategies in the development and use of CALL-software. Finally the results of a survey are presented in which language teachers presented their views on the 'context and conceptualization' of CALL. From this Levy draws the conclusion that some sort of CALL-theory is still missing and desirable.
Lieber, Gothild:
An international bibliography of computer-assisted language learning with annotations in German. 2. Edition (Jung, 1988). Frankfurt am Main : Lang, 1993
Ludewig, Petra:
Korpusbasiertes Kollokationslernen - Computer-Assisted Language Learning als prototypisches Anwendungsszenario der Computerlinguistik : Habilitationsschrift an der Universität Osnabrück, Fachbereich Sprach- und Literaturwissenschaften, 2003
Note: In diesem Buch wird das System LogoTax beschrieben, das es einem Fremdsprachenlerner ermöglicht, sein persönliches elektronisches Wörterbuch anzulegen. Dabei wird er insbesondere bei der Eintragung von Kollokationen unterstützt, da hierzu Beispielsätze für Kollokationen aus Korpora extrahiert werden.
Das Buch bietet einen zunächst einen Überblich über CALL bzw. ICALL. Anschließend werden die Grundlagen für die Behandlung von Kollokationen sowohl aus theoretisch-linguistischer als auch aus lexikographischer Perspektive erläutert.
Schließlich erfolgen didaktische Überlegungen und eine Beschreibung des Systems.
Maas, Utz:
Grundzüge der deutschen Orthographie. Tübingen : Niemeyer, 1992
Menzel, Wolfgang:
Modellbasierte Fehlerdiagnose in Sprachlehrsystemen. Tübingen : Niemeyer, 1992
Note: In diesem Buch beschreibt Wolfang Menzel sein Konzept einer modellbasierten Fehlerdiagnose. Dabei wird die Grammatik als ein Modell implementiert, in dem zum Parsing jede morphosyntaktische Eigenschaft eines Wortes ausspezifiziert wird und zur Kontrolle der Kongruenz und der Rektion mit jedem anderen relevanten Wort verknüpft wird. Daraus ergibt sich die Möglichkeit, umfassende Fehleranalysen und Diagnosen zu erreichen.
Mitton, Roger:
English Spelling and the Computer. London : Longman, 1996
Ritter, Markus:
Computer und handlungsorientierter Unterricht. Donauwoerth : Ludwig Auer, 1995
Rüschoff, Bernd ; Wolff, Dieter:
Fremdsprachenlernen in der Wissensgesellschaft: zum Einsatz der Neuen Technologien in Schule und Unterricht. Ismaning : Hueber, 1999
Schulmeister, Rolf:
Grundlagen hypermedialer Lernsysteme: Theorie - Didaktik - Design. München : Oldenbourg, 1997
http://www.izhd.uni-hamburg.de/paginae/Book/Frames/Start_FRAME.htmlNote: (checked 05.2004) The URL is a link to the English Version of this book. Relevant is especially chapter 6 about 'Intelligent Tutoring: What is ITS?'. Schulmeister looks at these systems not from a technical position but from a didactic position. Therefore he focuses on the relation between theories of learning and the possibilities of ITSs. The first part gives an introduction into the field describing the main components of an ITS. Then a comparison with expert systems follows. After that he presents a few cases of systems where looks at how good the systems actually model psychological theories of learning and tries to demonstrate that almost all ITS-research so far has failed at creating a new learning theory-oriented approach in ITS.
Storch, Günther:
Deutsch als Fremdsprache. Eine Didaktik. Stuttgart : UTB, 1999
Note: Dieses Buch stellt eine allgemeine Einführung in die Didaktik von Deutsch als Fremdsprache dar. Von besonderem Interesse ist hier, dass gerade nicht ein ausgeprägter Konstruktivismus vertreten wird. Besondere Elemente sind eher Aspekte wie Interkulturelles Lernen etc.
Swartz, M. L. ; Yazdani, Masoud (Ed.):
Intelligent Tutoring Systems for Foreign Language Learning. Berlin : Springer, 1992
Note: This is the first major publication on ICALL-systems especially from Europe. Almost all somewhat larger projects are contained in this book.
Tomita, M.:
Efficient Parsing of NL: A Fast Algorithm for Practical Systems. Boston, Ma : Kluwer, 1986
Yazdani, Masoud (Ed.):
Multilingual multimedia: bridging the language barrier with intelligent systems. Oxford : Intellect, 1993
Incollection
deSmedt, William H.: Herr Kommissar: An ICALL Conversation Simulator for Intermediate German. In: Holland, V.M. ; Kaplan, J.D. ; Sams, M.R. (Ed.) Intelligent Language Tutors. Mahwah, NJ : Erlbaum, 1995
Note: In this text the system 'Herr Kommissar' is presented. Through dialogs with the system the user is in charge of solving crimes. The main task consists of interrogating the suspects in order to identify the guilty person. The system has a knowledge management module and can therefore interpret the answers with respect to the knowlegde contained in the system. For example it can remind the user that a certain question has already been asked before. As the complete system is described the single modules are not described in much detail which makes it difficult to evaluate the concept of the program. This program was sold once for Apple-Computers but is apparently not available anymore.
Warschauer, Mark: Computer-assisted language learning: An introduction. In: Fotos, S. (Ed.) Multimedia language teaching. Tokyo : Logos International, 1996, pp 3-20
http://www.gse.uci.edu/markw/call.htmlNote: (checked 05.2004) This article provides brief overview of how computers have been used and are being used for language teaching. It focuses not on a technical description of hardware and software, but rather on the pedagogical questions that teachers have considered in using computers in the classroom. The introduction follows a historical path from 'behaviouristic CALL' to the Internet. The appendix provides a rather comprehensive typology of CALL programs and a list of further CALL resources. Note that this text is from 1996 and therefore some aspect such as constructionist approaches to language learning are missing. Nevertheless the paper gives a good introduction and overview from a certain perspective.
Inproceedings
Fouvry, Frederik: Robust Unification for Linguistics. In: ROMAND 2000 1st workshop on RObust Methods in Analysis of Natural language Data. Lausanne, 2000
Note: In this paper, an approach that is based on a more abstract and general level of robust parsing, viz. the formalism is presented. More concretely, a typed attribute logic (i.e. HPSG) is used based on Carpenter, (1992). In a first step it is ensured that the type hierarchy has certain properties to make a meaningful robustness possible, i.e. containing information about what went wrong. Then, a distance measure based on the logic is introduced to distinguish between different analyses. This paper describes a similar approach as Fouvry (2003) as it is also part of his dissertation.
Murphy, Maureen ; McTear, Michael: Learner Modelling for Intelligent CALL. In: Jameson, Anthony ; Paris, Cecile ; Tasso, Carlo (Ed.) User Modelling: Proceedings of the Sixth International Conference, UM97. Wien : Springer, 1997, pp 301-312
Note: This paper presents some result from the ReCALL project. The aim of this project was to work towards providing a more adequate and user-oriented interface for CALL. The adaptation was accomplished by developing a module that provides a parser-based analysis of the learner's response to the exercises of the program, a module that creates a model of the learner, and a module that controls the system's reactions to the learner's input and the structure of the materials offered to the learner.
Self, John A.: Knowledge, Belief and User Modelling. In: O'Shea, Tim ; Sgurev, Vasil (Ed.) Artificial Intelligence III: Methodology, Systems, Applications - Proceedings of the Third International Conference on Artificial Intelligence: Methodology, Systems, Applications (AIMSA '88). North-Holland, 1988, pp 3-9
Virvou, Maria ; Tsiriga, Victoria: Web Passive Voice Tutor: An Intelligent Computer Assisted Language Learning System over the WWW. In: Okamoto, T. ; Hartley, R. ; Kinshuk ; Klus, J.P. (Ed.) Proceedings IEEE International Conference on Advanced Learning Technology: Issues, Achievements and Challenges, 2001, pp 131-134
http://thalis.cs.unipi.gr/~vtsir/Virvou_Tsiriga_ICALT2001.pdfNote: (checked 05.2004) In this paper the so called Web Passive Voice Tutor (Web PVT), an adaptive web-based Intelligent Computer Assisted Language Learning (ICALL) program is described that is aimed at teaching non-native speakers the passive voice of the English language. Web PVT incorporates techniques from ITS and Adaptive Hypermedia technologies to provide students with individualized instruction and feedback. The system uses a combination of stereotypes and the overlay technique for the initialisation of the student model, which is then refined by observing the student while working with the system. The resulting student model is used for the annotation of the links to topics presented to the student.
However the techniques used in the system are only mentioned in the text and not explained.
Therefore only little information about the system is contained in the text as it is only described very briefly.
Zock, Michael: Sorry, but what was your name again, or, how to overcome the tip-of-the tongue problem with the help of a computer?. In: Proceedings of the COLING-Workshop on Building and Using Semantic Networks. Taipeh, 2002
Note: The paper presents a concept for the structuring of an electronic dictionary following especially psycholinguistic facts about access to the mental lexicon.
Work done by psychologists reveals that people being in this so called tip-of-the-tongue state (TOT) know a lot about the word : meaning, number of syllables, origine, etc. Speakers are generally able to recognize the word, and if they produce an erroneous word, that token shares many things with the target word (initial/final letter/phoneme, part of speech, semantic field, etc.). Three methods (access by form, access by meaning and a combination of both) that assist the speaker/writer by revealing the word that's on his/her mind (tongue/pen) are presented, the first one being implemented.
Internet
Wazel, Gerhard:
Evaluationskriterien für sprachliche Multimedia-Software, 2000
http://www.iik.com/theorie/theoretisches/kriterien.htmlNote: (checked 6.2004) Hierbei handelt es sich um einen AUszug aus einem umfangreicheren Papier. Der Kokus liegt einerseits auf allgemeinen Kriterien zur Bewertung von CALL-Software. Andererseits werden aber auch deutlich Empfehlungen gegeben, welche Fähigkeiten ein Programm besitzen sollte, ohne allerdings auf die technischen Leistungen einzugehen. Verschiedene Bereiche werden in geschachtelten Listen detailliert dargestellt.
Phdthesis
Heift, Trude:
Designed Intelligence: A Language Teacher Model. Burnaby, B.C., 1998
http://www.sfu.ca/langlab/trude/pubs.htmlNote: (checked 6.2004) This is the dissertation which presents the groundwork for the system 'German Tutor', later known as 'E-Tutor' which has been presented in numerous publications. It uses the grammar formalism HPSG in order to analyse erroneous sentences in a more or less non-anticipating way. The system is only able to detect very specific constituent errors, i.e. errors in linear precedence. Apart from recognizing errors it also has a module for disambiguating sentence readings and a student module for 'informed' feedback.
Reuer, Veit:
PromisD - Ein Analyseverfahren zur antizipationsfreien Erkennung und Erklärung von grammatischen Fehlern in Sprachlehrsystemen, 2003
Note: In dieser Dissertation wird das System PromisD beschrieben, das auf einigen Ideen des Studienprojekts Promise (Bauer 1994) aufbaut. Zunächst findet eine Einordnung des Systems innerhalb des Gebietes CALL und die Behandlung von fehlerhaften Äußerungen von Fremdsprachenlernern statt. Anschließend wird das von Reuer entwickelte System präsentiert, wobei der Schwerpunkt auf der Vorstellung des Fehler-sensitiven Parsing-Algorithmus liegt. Es wird versucht, jeweils ein Verfahren zum sensitiven Verfahren für die aus der LFG bekannten F- und C-Strukturen zu finden.
Proceedings
Techreport
Althaus, Nadja ; Beck, Kathrin ; Bennöhr, Jasmine ; Bernedo, Gordon ; Boeck, Manuel ; Elbers, Michael ; Kugel, Felix ; Scherbaum, Stefan ; Widdra, Tobias ; Wissmann, Jens:
Abschlussbericht des Studentenprojekts MAPA im Masterprogramm Cognitive Science. Osnabrück, 2003
Note: In the MAPA study project, students from the Universities of Osnabrück, Tübingen and Bochum participated in a joint effort to develop a framework that allows the mapping of knowledge in a cognitively adequate way. The approach connects to existing techniques of mind mapping, concept mapping and the like. A major assumption is that the internal mental representation of knowledge is networklike and that therefore its externalisation must be done accordingly. Due to the complexity of knowledge however, the external representations consist only of cues to knowledge rather than a representation of knowledge itself. The report reports on the foundations of mind mapping, presents the implemented system and has an extra chapter about an vocabulary trainer integrated into the framework (Beck 2003).
Bauer, Pascal ; John, Roul S. ; Kronenberg, Friedrich ; Krüger, Anja ; Menzel, Andre ; Reuer, Veit ; Unsöld, Robert:
Abschlussbericht PROMISE Studienprojekt. Osnabrück, 1994
Note: PROMISE is an acronym for "PROjekt Mediengestütztes Interaktives Sprachelnlernen Englisch" (i.e. media supported interactive language learning - English). In PROMISE, the communicative approach known from language pedagogy is realized by putting the learner in an adventure-game-like setting where authentic and purposeful language use is possible. Within a situational frame (e.g. a road accident) the student is guided through a series of dialogue exercises with "free" learner input in which the student "talks" with a simulated dialogue partner in the foreign language and gets meaningful responses from this partner. In order to realize this a number of modules need to interact generating the appropriate feedback especially with respect to language errors the learner makes. An LFG-parser is included as well as a module for handling the dialogue. Furthermore a world knowledge database allows among others the special treatment of spatial expressions. A language learner can therefore be supported when learning the usage of these patterns.
Bongardt, Gerd ; Dalinghaus, Klaus ; Dittman, Hendrik ; Huber, Michael ; Krumeich, Alexander ; Peylo, Christoph ; Rehm, Georg ; Rother, Andreas ; Teiken, Wilfried:
Semf-Abschlussbericht. Osnabrück, 1996
http://www.cogsci.uni-osnabrueck.de/~semf/Note: (checked 10.2001) In diesem Bericht werden die Ergebnisse des SEMF-Projekts vorgestellt. In diesem Projekt wurde versucht, Fehlleistungen eines Sprachlerners im semantischen Bereich zu erkennen und dazu Feedback zu bieten. Grundlage dafür ist eine so genannte Blocksworld, in der der Lerner Zustände beschreiben kann. Auf der Grundlage des vom System berechneten Zustandes der Welt kann eine entsprechende Rückmeldung gegeben werden.
Erpenbeck, Arno ; Koch, Britta ; Kummer, Norman ; Reuter, Philip ; Tschorn, Patrick ; Wagner, Joachim:
KOKS-Abschlussbericht. Osnabrück, 2001
http://www.cogsci.uni-osnabrueck.de/~koks/Note: In der ersten Phase wurden zur Indentifizierung der Kollokationen mit Hilfe eines Taggers und eines bilingualen Lexikons zunächst die Sätze alignet. Basierend auf einer statistischen Analyse der hochfrequenten POS-Abfolgen in den Korpora wurden Chunking-Regeln entwickelt, die in einer zweiten Phase ein Phrasen-Alignment ermöglichen. Darauf aufbauend wurde in einer dritten Phase das bilinguale Lexikon herangezogen, um Phrasen zu identifizieren, deren alignetes Pendant nicht durch die Übersetzungen im Lexikon abgedeckt wird. Damit wurde ein Verfahren zur Identifikation von Kollokationen entwickelt, das sich besonders durch die Berücksichtigung linguistischer Kriterien in der Definition von Kollokationen auszeichnet.
Krüger, Anja ; Dittmann, Hendrik ; Murphy, Maureen:
Grammar Based Error Diagnosis in CALL, 1997
Note: Informatics Research Reports - University of Ulster. This report is probably only available by contacting one of the authors. This report presents the approach taken in the ReCALL project, in which a large coverage parser is extended to include rules for the semantic analysis of certain constructions. It by no means a general approach as every information has to be entered into the grammar the same way it has to be added to a knowledge base. However for certain situations this may be a feasible approach.
Sagvall Hein, Anna:
A Grammar Checking Module for Swedish (SCARRIE Deliverable 6.6.3). Uppsala, 1998
http://www.ling.uu.se/wp/wp12c.pdfNote: (checked 6.2004) This report from the SCARRIE project presents the approach taken for grammar checking. It uses a bottom-up chart-parser (the Uppsala Chart Processor (UCP)) to generate as much chart items as the grammar allows. A second module then scans the chart with the help of error-rules in order to detect grammar errors. The system is not evaluated against a corpus but the report includes a detailed description of the errors covered.