0264 A language usage-based service for providing formative feedback and learner positioning


13:00 - 14:00 on Wednesday, 8 September 2010 in Pos


0264 A language usage-based service for providing formative feedback and learner positioning
Gaston Burek, Gillian Armitt, Dale Gerdemann, Bernhard Hoisl, Robert Koblischke, Christoph Mauerhofer, Petya Osenova, Kiril Simov


0264 A language usage-based service for providing formative feedback and learner positioning
Gaston Burek, Gillian Armitt, Dale Gerdemann, Bernhard Hoisl, Robert Koblischke, Christoph Mauerhofer, Petya Osenova, Kiril Simov
Self-directed learners can benefit from personalised 'feedback on demand' during their learning but this is often not practical owing to tutors' time constraints. Tutors can benefit from computerised support with positioning learners and providing individualised feedback at the right level. Language technologies (phrase extraction and Latent Semantic Analysis (LSA)) analysing speech genres (Bakhtin 1986) offer an opportunity to address these issues. Experts in a domain develop a speech genre, with characteristic phrases. Becoming an expert involves the adoption of the community's speech genre. Thus, learner knowledge can be evaluated by means of textual distance based on characteristic phrases. Through the EU-funded LTfLL project, bespoke on-line software was designed as Service Usage Models in collaboration with a commercial training partner. Ontology and lexical resources are used to identify concepts covered by learners. Positioning is achieved through combining linguistic patterns (phrases) extraction and LSA (Burek and Gerdemann 2009) to compare learner texts with a corpus of expert texts built from IT learning materials and high quality learner texts. The software indirectly measures learners' degree of expertise by textual distance to the relevant speech genre. The poster illustrates the feedback and positioning information provided by the user interface and how the language technology-based comparison of learner and expert texts is achieved to compare speech genres. The poster illustrates learner and tutor views of the software. Learners receive formative feedback on the strong and weak areas of texts submitted to the system, to help them revise their texts before submission to tutors for positioning. Tutors receive a system-generated provisional grading with formative feedback and can adjust the grading and feedback. A commercial IT training company is piloting the software for short IT courses. Validation results are presented demonstrating the value of the software for supporting self-directed learning and enabling tutors to provide personalised feedback. The software facilitates more effective learning and improves tutor efficiency through informing personalised feedback. The software is applicable to other domains through changing the expert texts to the relevant domain. Work is under way to extend the service to Medicine.