Presentation on theme: "Web Passive Voice Tutor: an Intelligent Computer Assisted Language Learning System over the WWW Maria Virvou & Victoria Tsiriga Department of Informatics,"— Presentation transcript:
Web Passive Voice Tutor: an Intelligent Computer Assisted Language Learning System over the WWW Maria Virvou & Victoria Tsiriga Department of Informatics, University of Piraeus, 80 Karaoli & Dimitriou St.,Piraeus 18534, Greece
Overview Web-based Educational Systems Web Passive Voice Tutor Architecture of Web PVT Functionality of Web PVT Adaptive Navigation Support Intelligent Analysis of Solutions Conclusions and Future Work
Web-based Educational Systems Recently, a lot of research energy is put on Web-based instruction. Benefits of Web-based education are independence of teaching and learning with respect to time and space. However, most WWW learning systems, lack the capabilities of individualised learning support. This is even more the case for Web-based language tutoring systems.
Web Passive Voice Tutor (Web PVT) Web PVT is a Web-based Intelligent Computer Assisted Language Learning (ICALL) system aimed at teaching non-native speakers the passive voice of the English language. The adaptivity of the system is implemented by presenting students with different, dynamically constructed HTML pages. The decision about the content of each page is based on the model of each individual student.
Architecture of Web PVT SERVER Domain KnowledgeStudent Modelling Tutor CLIENT Web Browser (Dynamic HTML Pages) Internet
Architecture of Web PVT Web PVT represents its domain knowledge using a semantic net depicting the relations between the domain concepts. There are three kinds of links between nodes: part-of, is-a, and prerequisite. Student modelling is based on a combination of stereotypes and the overlay technique. Stereotypes are used to initialise the student model. The system then updates the model based on the actions of the student. The tutoring component makes the pedagogical decisions, based on the strengths and weaknesses of a student. The user interface consists of a set of dynamically constructed HTML pages and forms.
Functionality of Web PVT Web PVT incorporates techniques from Adaptive Hypermedia and Intelligent Tutoring Systems: to support the navigation of the student through the course material, to perform intelligent analysis of the students answers, to tailor the feedback to errors to each individual student.
Adaptive Navigation Support Web PVT uses adaptive link annotation to support the navigation of the student through the course material. The idea of adaptive annotation is to augment links with some form of comments that can tell the user more about the current state of the nodes behind the links. Web PVT uses different font types and icons to annotate the links of the theory hyperdocument. The system distinguishes between four possible states of links: highly recommended, ready and recommended, visited and known and not ready.
Adaptive Navigation Support
Intelligent Analysis of Solutions When a student has solved an exercise, her/his answer is analyzed. If it is correct, the student is congratulated. In case of an erroneous answer, the system performs error diagnosis. Sometimes, a mistake of a student may be attributed to more than one categories of error. In such cases, Web PVT resolves the ambiguity taking into account the individual features of the user that have been recorded in previous interactions.
Intelligent Analysis of Solutions (Example) If a student has typed the word teacher instead of teachers in a sentence converted to the passive voice, this could either be an accidental slip or a singular/plural mistake. If the particular student has not previously made singular/plural mistakes but has made carelessness mistakes then the system favours carelessness as the most probable cause of the mistake.
Conclusions and Future Work In this talk we have described Web PVT, a web- based ICALL system. The system uses the link annotation technique so as to provide adaptive navigation support. Web PVT is capable of performing error diagnosis and ambiguity resolution based on the long term student model. Within the future plans is the improvement of the student modelling component, so that it may cope with temporal aspects of the learning process.