REAP.PT (Computer Aided Language Learning - Reading Practice)

From HLT@INESC-ID

Funded by: FCT (Carnegie Mellon | Portugal Research Projects Program) [CMU-PT/HuMach/0053/2008]
Start date: 01 April 2009
Duration: 36 months

Summary

The main purpose of this project is to build a Portuguese version of REAP (“REAding Practice”), a tutoring system developed at LTI to support the teaching of a language for either native or foreign speakers, through the activity of reading and focusing the students in learning vocabulary in context.


Partners

  • INESC-ID – Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento em Lisboa, Portugal
  • Carnegie Mellon University (CMU)
  • Universidade do Algarve (UAlg)
  • Fundação da Universidade de Lisboa (FUL/UL)


Main Researchers

INESC-ID UAlg FUL/UL LTI/CMU
Nuno Mamede Jorge Baptista Maria do Céu Viana Maxine Eskenazi
Isabel Trancoso Maria de Lurdes Cabral Palmira Marrafa Adam Skory
David Martins de Matos Joaquim Guerra Ana Isabel Mata Gabriel Parent
Thomas Pellegrini Neuza Baptista
Luís Marujo
Rui Correia
Ana Cristina Mendes
Ricardo Ribeiro
Rui Amaral
José David Lopes

Description

The development of REAP.PT has started very recently in the framework of the CMU-Portugal program. A rudimentary version is already in place, resulting from the tight cooperation of the LTI and L2F teams. The current proposal involves two major phases: in the first one, the project teams will build a baseline version for European Portuguese, integrating all the linguistic tools and resources for that language, in order to build an equivalent version to the current English system, and making the necessary adaptations for this typologically different language. In the second phase, the project teams will concentrate on the very wide range of open research challenges presented by the possibility of learning from current texts on topics of the student´s interests. Therefore, in most of the 9 tasks of this project, two significant milestones can thus be identified, corresponding to the development of the baseline and extended versions.

The project is structured into 9 main tasks: corpus collection (T1), readability (T2), interface (T3), definitions (T4), search (T5), question generation (T6), student modeling (T7), and oral comprehension (T8). The last task (T9) will report field trials.

Task 1 (Corpus Collection) will be responsible for crawling the web for Portuguese pages and filtering them, keeping only those that satisfy all the requirements. The Readability task (Task 2) will attribute a reading difficulty value to each document collected in the previous task. Task 3 (Interface) will adapt the REAP interface to Portuguese, and Task 4 (Definitions) will integrate a Portuguese dictionary in the system. Task 5 (Search) is responsible for the retrieval of texts satisfying particular pedagogical constraints such as reading level and text length. Task 6 (Question Generation) will automatically generate fill-in-the-blank questions with multiple choice, which will be used as practice exercises for focus words. Following the practice session, the system updates the student model, defined on task 7, based on his or her performance. Task 8 will focus on practicing the oral comprehension and Task 9 will evaluate the several versions of the system.


Publications

Luis Marujo, José Lopes, Nuno J. Mamede, Isabel Trancoso, Juan Pino, Maxine Eskenazi, Jorge Baptista and Céu Viana, Porting REAP to European Portuguese, ISCA International Workshop on Speech and Language Technology in Education (SLaTE 2009), Sep. 2009 , ISCA.

Jorge Baptista, Neuzza Costa, Joaquim Guerra, Marcos Zampieri, Maria Lurdes Cabral and Nuno J. Mamede, P-AWL Academic Word LIst for Portuguese, PROPOR 2010, Porto Alegre, Brazil, April 2010. [to appear]

Results

Portuguese Academic Word List (P-AWL) [available in March/2010]