Seminar für Sprachwissenschaft

Projekte Theoretische Computerlinguistik

SFB 833 - Projekt A4 "Bedeutungsvergleich im Kontext"

Projektleiter(in): Detmar Meurers
Mitarbeiter: Kordula De Kuthy, Researcher
Björn Rudzewitz, Researcher
Homepage: http://purl.org/dm/projects/sfb833-a4/
Absicht :  
Motiviert durch die generelle Frage, wie Bedeutungsvergleiche auch in realistischen Situationen erfolgen können, in denen nicht wohlgeformte Sprache oder Unterschiede im situativen Wissen oder Weltwissen eine vollständige Analyse erschweren oder unmöglich machen, untersucht das Projekt, welche linguistischen Repräsentationen für einen computerlinguistischen Vergleich der Bedeutung von Sätzen und Textfragmenten effektiv und robust zu verwenden sind und für welche Aufgaben und Kontexte sich solche Vergleiche mit einer flachen semantischen Analyse berechnen lassen.

KANSAS

Kontakt:

Schröter, Hannes, PD Dr.,  Projektkoordination, wiss. Leitung

Beteiligte Personen:

Becker-Mrotzek, Michael, Prof. Dr. (wiss. Leitung)

Galasso, Sabrina (Wissenschaftliche Bearbeitung)

Geppert, Theresa (Wissenschaftliche Bearbeitung)

Jambor-Fahlen, Simone , Dr. (Wissenschaftliche Bearbeitung)

Meurers, Detmar, Prof. Dr. (wiss. Leitung)

Schrader, Josef, Prof. Dr. (wiss. Leitung)

Schröter, Hannes, PD Dr. (Projektkoordination, wiss. Leitung)

Steinborn, Waltraud (Mitarbeit)

Triulzi, Marco (Mitarbeit)

Weiß, Zarah Leonie (Wissenschaftliche Bearbeitung)

Homepage:

https://www.die-bonn.de/id/35640/about/html/ 

  https://www.die-bonn.de/kansas/default.aspx 

September 2017 bis August 2020

FRAGESTELLUNG

Wie können Lehrkräfte in Kursen der Alphabetisierung und Grundbildung sowie Deutsch als Zweitsprache bei der Recherche und Auswahl von Sprachlerntexten im Internet und in elektronischen Textsammlungen unterstützt werden, die sowohl dem gegenwärtigen individuellen Kompetenzniveau als auch den thematischen Interessen der Kursteilnehmenden entsprechen?

PROJEKTZIEL

In diesem interdisziplinären Projekt von Erwachsenenbildung, Computerlinguistik, Fachdidaktik und Kognitionspsychologie soll eine kompetenzadaptive, nutzerorientierte Suchmaschine für authentische Sprachlerntexte (KANSAS) entwickelt, erprobt und evaluiert werden.

KANSAS soll primär Lehrkräfte in Kursen der Alphabetisierung und Grundbildung dabei unterstützen, Sprachlernmaterialien im Internet und in elektronischen Textsammlungen zu recherchieren, die sowohl einer dem gegenwärtigen individuellen Kompetenzniveau angemessenen sprachlichen Komplexität als auch den thematischen Interessen der Kursteilnehmenden entsprechen. Eine zweite Adressatengruppe bilden Lehrkräfte für Deutsch als Zweitsprache, die beispielsweise in Integrationskursen tätig sind.

Die Entwicklung einer zusätzlichen, vereinfachten Version von KANSAS soll fortgeschrittene Lernende dazu in die Lage versetzen, Texte mit geeigneter sprachlicher Komplexität selbständig zu recherchieren. Dies soll den Lernenden sowohl eine eigenverantwortliche Vertiefung von erlernten Kompetenzen als auch einen erleichterten Zugang zur digitalen Informationswelt ermöglichen.

ICALL

Projektleiter(in): Detmar Meurers
Homepage:

http://www.sfs.uni-tuebingen.de/icall/ 

Absicht :  

Research in Intelligent Computer-Assisted Language Learning (ICALL) integrates Natural Language Processing into Computer-Assisted Language Learning. For such research to be innovative and sustainable, we believe that it needs to combine Second Language Acquisition (SLA) research, precise linguistic modeling, and sound computational linguistic methods (Meurers 2012Meurers & Dickinson 2017). On this foundational research, we ground the development of digital tools addressing real-life education needs, in school and lifelong learning.

FeedBook

Projektleiter(in): Detmar Meurers
Mitarbeiter:

Dr. Kordula De Kuthy, Researcher
Björn Rudzewitz, Researcher

Ramon Ziai, Researcher

Homepage:

http://feedbook.website 

Absicht :  

Das FeedBook ist ein interaktives webbasiertes Workbook für den Englischunterricht an Gymnasien. Es wird von Computerlinguisten der Universität Tübingen in Zusammenarbeit mit Diesterweg Verlag auf Basis des aktuellen Workbooks Camdentown entwickelt.

Hauptziel bei dieser webbasierten Version des Workbook ist es, den SchülerInnen individuelles, interaktives Feedback bei der Bearbeitung der Aufgaben zu liefern. Dabei werden aktuelle computerlinguistische Methoden zur Analyse von Form und Bedeutung im Aufgabenkontext eingesetzt und für das FeedBook weiterentwickelt.

Eine erste Version des FeedBooks ist bereits im Einsatz an 4 Schulen. Es umfasst insgesamt 55 Aufgaben, darin enthalten sind Lückentextaufgaben und Leseverständnisaufgaben, die den Lehrstoff des gesamten 7. Schuljahres and Gymnasien abdecken.

TAGARELA

Projektleiter(in): Detmar Meurers
Homepage:

http://sifnos.sfs.uni-tuebingen.de/tagarela/index.py/main

About:  

Teaching Aid for Grammatical Awareness, Recognition and Enhancement of Linguistic Abilities

TAGARELA is an intelligent computer-assisted language learning (ICALL) system designed to be fully integrated into the Portuguese Individualized Instruction Program at The Ohio State University. Originally created at OSU, it is now further developed at Tübingen University and employed in Portuguese courses at UMASS Amherst. TAGARELA can be viewed as an intelligent automatic workbook that provides students with opportunities to practice their reading, listening, and writing skills. Because it is a web-based system, it can be used anywhere there is a computer with internet access.

TAGARELA uses state-of-the-art Natural Language Processing (NLP) technology to analyze students' input, and detect spelling, morphological, syntactic and semantic errors. Its NLP components allow the system to handle activities that go beyond the usual multiple choice or fill-in-the-blanks used by regular CALL systems. Its exercise types are similar to the ones typically found in current workbooks, such as:

  • listening and reading comprehension
  • picture description
  • vocabulary practice
  • phrasing and re-writing

 

TAGARELA has a student model that keeps track of each student's individual performance, and an instructor model that carries information about activity and error types. Together the student and instructor models allow the system to choose the best feedback strategy to use with each individual learner based on the level of the activity, type of task, characteristics of error, and learner profile.

At Tübingen University, TAGARELA was subsequently re-implemented in the UIMA framework. The new version retains the full functionality of the first TAGARELA and offers the following new capabilities:

 

  • True extensible annotation-based processing
  • Adaptation to activities based on information from the activity model
  • Adaptation of feedback for individual students based on basic student modeling

The Ohio State University has the longest running individualized language programs in the country. With TAGARELA, OSU is the first university to have an ICALL system with NLP technology specifically designed to fulfill the needs of students in such language programs. The system was successfully used by the students in individualized instruction and regular Portuguese courses at OSU in Spring 2007. 

This project was funded 2005-2007 by an Arts and Humanities Grant for Innovation under the project title "Bridging the Gap between Research in Natural Language Processing and Individualized Language Instruction."

FLAIR

Projektleiter(in): Detmar Meurers
Mitarbeiter:

Maria Chinkina,

Madeeswaran Kannan

Homepage:

http://sifnos.sfs.uni-tuebingen.de/FLAIR/ 

Absicht :  

FLAIR is an online tool for language teachers and learners that: 

  • ● searches the web for a topic of interest 
  • ● analyzes the results for grammatical constructions and readability levels 
  • ● re-ranks the results according to your (pedagogical or learning) needs specified in the settings

WERTi

Projektleiter(in): Detmar Meurers
Homepage:

http://sifnos.sfs.uni-tuebingen.de/WERTi/ 

About:  

WERTi is an intelligent computer-assisted language learning (ICALL) system designed to provide supplementary language learning activity resources to ESL students. It can be viewed as an intelligent automatic workbook, providing an unlimited number of activities designed to foster awareness of English grammatical forms and functions. Because it is a web-based system, it can be used anywhere there is a computer with internet access.

 

The WERTi activities are automatically derived from authentic English texts, obtained from any web page of interest. The task of retrieving texts appropriate for learners is handled by a separate Information Retrieval for Language Learning (IR4LL) project.

WERTi uses state-of-the-art Natural Language Processing (NLP) technology to generate exercises, identifying targeted lexical and phrasal material through a combination of tokenization, lemmatization, morphological analysis, part-of-speech tagging, and shallow parsing. For the types of activities currently being implemented, the focus in WERTi is on creative and efficient use of existing freely available NLP technology.

The activity types WERTi provides follow a pedagogically motivated progression from receptive presentation, to productive presentation, to controlled practice. Specifically, activities include the coloring of targeted forms, having the learner find and click on targeted forms, and finally, controlled practice activities such as fill-in-the-blank or editing tasks. WERTi provides these activity progressions for a variety of grammar topics.

The Education First-Cambridge Learner Corpus: a data-driven approach to second language learning

Projektleiter(in):   
Mitarbeiter: Detmar Meurers 
Homepage:

http://www.ling.cam.ac.uk/ef-unit/

Absicht :
 
Prof. Meurers collaborates with the Research Center for English and Applied Linguistics on the "Education First-Cambridge Learner Corpus: a data-driven approach to second language learning". 

VIEW

Projektleiter(in): Detmar Meurers 
Mitarbeiter: Adriane Boyd, Researcher 
Homepage:

http://purl.org/icall/view    

http://www.sfs.uni-tuebingen.de/%7Edm/projects.html

Absicht :  
Multilingual Platform for the European Reference Levels: Interlanguage Exploration in Context EU Lifelong Learning project 518989-LLP-1-2011-1-DE-KA2-KA2MP