ERROR ANALYSIS AND HANDLING IN ARABIC ICALL SYSTEMS Khaled F. Shaalan 12 Habib E. Talhami 12 Institute of Informatics 1 The British University in Dubai P. O. Box 502216, Dubai,UAE 2 Honorary Fellow, School of Informatics, University of Edinburgh {khaled.shaalan,habib.talham}@buid.ac.ae ABSTRACT Arabic is a Semitic language that is rich in its morphology and syntax. The very numerous and complex grammar rules of the language could be confusing even for Arabic native speakers. Many Arabic intelligent computer- assisted language-learning (ICALL) systems have neither deep error analysis nor sophisticated error handling. In this paper, we report an attempt at developing an error analyzer and error handler for Arabic as an important part of the Arabic ICALL system. In this system, the learners are encouraged to construct sentences freely in various contexts and are guided to recognize by themselves the errors or inappropriate usage of their language constructs. We used natural language processing (NLP) tools such as a morphological analyzer and a syntax analyzer for error analysis and to give feedback to the learner. Furthermore, we propose a mechanism of correction by the learner, which allows the learner to correct the typed sentence independently. This will result in the learner being able to figure out what the error is. Examples of error analysis and error handling will be given and will illustrate how the system works. KEY WORDS Arabic ICALL, Error Analysis, Error Handling, NLP- based intelligent feedback 1. Introduction Computer-assisted language learning (CALL) addresses the use of computers for language teaching and learning. CALL emerged in the early days of computers. Since the early 1960's, CALL systems have been designed and built. The effectiveness of CALL systems has been demonstrated by many researchers [1] [2]. More than a decade ago, Intelligent Computer-Assisted Language Learning (ICALL) started as a separate research field, when artificial intelligence (AI) technologies were mature enough to be included in language learning systems. The beginning of the new research field was characterized by intelligent tutoring systems (ITS), which embedded some NLP features to extend the functionality of traditional language learning systems. The continuous advances in ICALL systems have been documented in several publications [3] [4] [5] [6]. One of the weaknesses of current Arabic ICALL systems is that learners cannot key in Arabic sentences freely. Similarly, the system cannot guide the learner to correct the most likely ill-formed input sentences. The learner just accepts the information, which has been pre- programmed into the system. For these systems to be useful, more research to combine NLP techniques with language learning systems is needed [7]. Parsing, the core component in ICALL systems, allows the system both to analyze the learner’s input and to generate responses to that input [8]. Allowing learners to phrase their own sentences freely without following any pre-fixed rules can improve the effectiveness of ICALL systems, especially when the expected answers are relatively short and well- focused [9]. Both the well- and ill-formed structure of the input sentence can be recognized. The learner should be allowed to correct the typed sentence independently. This paper describes error analysis and handling in an Arabic ICALL system using NLP techniques, which is a step towards enhancing current Arabic ICALL systems. The current system guides learners to recognize by themselves the errors or improper usage of their language constructs. In other words, it helps learners to learn from their own mistakes. It doesn't give them the correct answer directly but it enables them to try over and over again. In this system, we use NLP tools such as a morphological analyzer, a syntax analyzer, and an error analyzer to give feedback to the learner. Furthermore, we propose a mechanism of correction by learners which allows the learner to correct the typed sentence independently. The rest of the paper is organized as follows: Arabic ICALL framework is summarized in section 2. This is followed by a description of the Arabic sentence analysis in Section 3. Next, the proposed feedback component responsible for error analysis and error handling is described in Section 4. Finally, a conclusion and recommendations for further enhancements are given in section 5. 502-141 109