Article Preview
TopIntroduction
Practice and research in the composition education using computer and network have been more and more active (Thompson, 1993; Sugimoto, 2004). Compared with traditional paper-writing works, online composition submission system can save a lot of time and effort. While for the normal correction functions in many online composition submission systems, student’s learning stage is usually not considered, and there is not enough correction tools supplied to teachers corresponding to a variety of misuse. The composition and correction information has often been accumulated respectively and not easily been retrieved. (Urasaki & Kogo, 1998; Kawai, 1996). One more important issue is that through online composition submission system, large amount of written texts produced by students and teachers can be collected. This kind of information is called a learner corpus (the plural is learner corpora), which is very important in second language education because the specific learning situations of learners can be closely analyzed (Granger, 2002; Granger, 2009; McEnery & Wilson, 2001). However, there is still little effective application on how to utilize the collected learner corpus in pedagogy (Nesselhauf, 2004). For example, a language learning environment named Intelligent Web-based Interactive Language Learning (IWiLL) has been constructed in Tamkang University, which allows both teachers and students to create and search an online database of student essays and teacher annotations (Wible, Huo, Chien, Liu & Tsao, 2001). In Tokushima University of Japan, a computer mediated language learning environment called CoCoA (Communicative Correction Assisting System) that supports foreigners and teachers to exchange corrected Japanese compositions via Internet has been created. They also have proposed CCML (Communicative Correction Markup Language) to record and exchange corrected compositions with some marks and some comments (Yano, Ogata & Sakakibara, 1997). In these two projects, learner corpora are constructed timely through e-Learning platform; however, there is no specific proposal on the pedagogical applications of these constructed learner corpora.
In this research, we have constructed a web-based composition/correction system as part of Tutorial Chinese Platform (TCP) (Liu, Scott, Kato, Kato & Urano, 2011) in Waseda University. The system will not only suitable for students in writing compositions, but also support teachers in correcting and making comments on compositions, as well as in retrieving and detecting students’ common misuses (Liu, Sunaoka & Urano, 2006). Moreover, the compositions from students, together with the correction information from teachers, has been structured and constructed automatically into learner corpora, and the retrieval function has also been developed and combined with another CALL (Computer Assisted Language Learning) drill system to create more effective exercises according to students’ misuses (Liu & Urano, 2011; Liu, Liu & Urano, 2012).
This research proposes a new mechanism to integrate both construction and application of learner corpus through the combination of web based composition/correction system and CALL drill creation function. Automatically XML tagging for composition/correction corpus and corpus retrieval function play crucial roles in this mechanism. It makes this system different to previous ones that not only can construct learner corpora timely, but also can apply these corpora in pedagogical activities effectively. This system has been carried out for years in Waseda University and the vitality of its role has been confirmed by the improvement in students’ language skills and the positive results of its usability evaluation.