Russian Morphological Processing for ICALL System architecture - PowerPoint PPT Presentation

Russian Morphological Processing for ICALL Introduction & Motivation ICALL context Russian Morphological Processing for ICALL System architecture Exercise design Error types Morphological Markus Dickinson and Joshua Herring analysis Lexicon Dept. of Linguistics, Indiana University Error detection Constructing the Lexicon ACL Workshop on Building Educational Applications Columbus, OH Summary & June 19, 2008 Outlook References 1 / 20

Introduction & Motivation Russian Morphological Processing for ICALL Intelligent computer-aided language learning (ICALL) Introduction & systems are ideal for language pedagogy Motivation ◮ provide additional practice outside classroom ICALL context System ◮ aiding awareness of language forms & rules (see Amaral architecture Exercise design and Meurers 2006) Error types Morphological analysis Lexicon Error detection Constructing the Lexicon Summary & Outlook References 2 / 20

Introduction & Motivation Russian Morphological Processing for ICALL Intelligent computer-aided language learning (ICALL) Introduction & systems are ideal for language pedagogy Motivation ◮ provide additional practice outside classroom ICALL context System ◮ aiding awareness of language forms & rules (see Amaral architecture Exercise design and Meurers 2006) Error types Morphological However: analysis Lexicon ◮ Few ICALL systems in existence today Error detection ◮ German (Heift and Nicholson 2001) Constructing the Lexicon ◮ Portuguese (Amaral and Meurers 2006, 2007) Summary & ◮ Japanese (Nagata 1995) Outlook References 2 / 20

Introduction & Motivation Russian Morphological Processing for ICALL Intelligent computer-aided language learning (ICALL) Introduction & systems are ideal for language pedagogy Motivation ◮ provide additional practice outside classroom ICALL context System ◮ aiding awareness of language forms & rules (see Amaral architecture Exercise design and Meurers 2006) Error types Morphological However: analysis Lexicon ◮ Few ICALL systems in existence today Error detection ◮ German (Heift and Nicholson 2001) Constructing the Lexicon ◮ Portuguese (Amaral and Meurers 2006, 2007) Summary & ◮ Japanese (Nagata 1995) Outlook ◮ Processing of ill-formed learner text focuses on a limited References set of languages and language types ◮ See Vandeventer Faltin (2003) and references therein 2 / 20

Introduction & Motivation Russian Morphological Processing for ICALL Intelligent computer-aided language learning (ICALL) Introduction & systems are ideal for language pedagogy Motivation ◮ provide additional practice outside classroom ICALL context System ◮ aiding awareness of language forms & rules (see Amaral architecture Exercise design and Meurers 2006) Error types Morphological However: analysis Lexicon ◮ Few ICALL systems in existence today Error detection ◮ German (Heift and Nicholson 2001) Constructing the Lexicon ◮ Portuguese (Amaral and Meurers 2006, 2007) Summary & ◮ Japanese (Nagata 1995) Outlook ◮ Processing of ill-formed learner text focuses on a limited References set of languages and language types ◮ See Vandeventer Faltin (2003) and references therein ⇒ Should expand to more language families 2 / 20

Re-usability Russian Morphological Processing for ICALL Introduction & Motivation Significant overhead in developing an ICALL system ICALL context System architecture Exercise design Error types Morphological analysis Lexicon Error detection Constructing the Lexicon Summary & Outlook References 3 / 20

Re-usability Russian Morphological Processing for ICALL Introduction & Motivation Significant overhead in developing an ICALL system ICALL context System architecture Exercise design Effort in producing an ICALL system can be reduced by: Error types ◮ reusing system architecture Morphological analysis ◮ evaluating and optimizing the architecture Lexicon Error detection Constructing the Lexicon Summary & Outlook References 3 / 20

Re-usability Russian Morphological Processing for ICALL Introduction & Motivation Significant overhead in developing an ICALL system ICALL context System architecture Exercise design Effort in producing an ICALL system can be reduced by: Error types ◮ reusing system architecture Morphological analysis ◮ evaluating and optimizing the architecture Lexicon Error detection ◮ adapting existing NLP tools Constructing the Lexicon ◮ and/or developing resource-light technology Summary & Outlook References 3 / 20

Re-usability Russian Morphological Processing for ICALL Introduction & Motivation Significant overhead in developing an ICALL system ICALL context System architecture Exercise design Effort in producing an ICALL system can be reduced by: Error types ◮ reusing system architecture Morphological analysis ◮ evaluating and optimizing the architecture Lexicon Error detection ◮ adapting existing NLP tools Constructing the Lexicon ◮ and/or developing resource-light technology Summary & Outlook It is important to determine where and how reuse of References technology is appropriate 3 / 20

Russian ICALL Russian Morphological Processing for ICALL We are developing an ICALL system for beginning learners of Introduction & Motivation Russian ICALL context System architecture Exercise design Error types Morphological analysis Lexicon Error detection Constructing the Lexicon Summary & Outlook References 4 / 20

Russian ICALL Russian Morphological Processing for ICALL We are developing an ICALL system for beginning learners of Introduction & Motivation Russian ICALL context ◮ Based on the TAGARELA system for Portuguese System architecture Exercise design (Amaral and Meurers 2006, 2007) Error types ◮ Q1: How can the technology in TAGARELA can be Morphological analysis adapted for efficient & accurate use with Russian? Lexicon Error detection Constructing the Lexicon Summary & Outlook References 4 / 20

Russian ICALL Russian Morphological Processing for ICALL We are developing an ICALL system for beginning learners of Introduction & Motivation Russian ICALL context ◮ Based on the TAGARELA system for Portuguese System architecture Exercise design (Amaral and Meurers 2006, 2007) Error types ◮ Q1: How can the technology in TAGARELA can be Morphological analysis adapted for efficient & accurate use with Russian? Lexicon Error detection ◮ Requires development of techniques to parse ill-formed Constructing the input for a morphologically-rich language Lexicon ◮ Q2: What kind of processing do we need, and are Summary & Outlook existing NLP tools reusable for this purpose? References 4 / 20

Russian ICALL Russian Morphological Processing for ICALL We are developing an ICALL system for beginning learners of Introduction & Motivation Russian ICALL context ◮ Based on the TAGARELA system for Portuguese System architecture Exercise design (Amaral and Meurers 2006, 2007) Error types ◮ Q1: How can the technology in TAGARELA can be Morphological analysis adapted for efficient & accurate use with Russian? Lexicon Error detection ◮ Requires development of techniques to parse ill-formed Constructing the input for a morphologically-rich language Lexicon ◮ Q2: What kind of processing do we need, and are Summary & Outlook existing NLP tools reusable for this purpose? References ◮ Q2a: What is the context for processing (i.e., the exercise requirements)? 4 / 20

Russian ICALL Russian Morphological Processing for ICALL We are developing an ICALL system for beginning learners of Introduction & Motivation Russian ICALL context ◮ Based on the TAGARELA system for Portuguese System architecture Exercise design (Amaral and Meurers 2006, 2007) Error types ◮ Q1: How can the technology in TAGARELA can be Morphological analysis adapted for efficient & accurate use with Russian? Lexicon Error detection ◮ Requires development of techniques to parse ill-formed Constructing the input for a morphologically-rich language Lexicon ◮ Q2: What kind of processing do we need, and are Summary & Outlook existing NLP tools reusable for this purpose? References ◮ Q2a: What is the context for processing (i.e., the exercise requirements)? ◮ Q2b: What are the expected types of morphological errors? 4 / 20

System architecture Russian Morphological Processing for ICALL From TAGARELA, we retain: Introduction & ◮ Modular separation of activities from analysis Motivation ◮ Each activity type has own directory, to ease in: ICALL context System ◮ loading different kinds of external files (e.g., sound) architecture Exercise design ◮ calling different processing tools (Amaral 2007) Error types Morphological analysis Lexicon Error detection Constructing the Lexicon Summary & Outlook References 5 / 20