Supporting the development of synchronous text-based computer-mediated communication with an intelligent diagnosis tool

To date, a look at the scientific literature on the construction and use of synchronous computer-mediated communication (CMC) support environments reveals that most researchers have focused either on exchanging information or on constructing and presenting posts. In this work, an intelligent collaborative synchronous CMC platform that detects whether the learners address the expected discussion issues is proposed. The concept maps related to the learning topics are first outlined by the instructor. After each learner issues a post on the synchronous CMC platform, a feature selection approach is adopted to derive the input parameters of a one-class Support Vector Machines (SVMs) classifier. The classifier then determines if the learners’ posts are related to the concept maps previously outlined by the instructor. Meanwhile, learner peers from the same group are asked to provide comments on the synchronous CMCs, and a group grading module is established in this work to evaluate the quality of the synchronous CMCs. If the evaluation results from the classifier and the group grading module are inconsistent, the instructor or the teaching assistant is consulted to verify the evaluation results. Notably, a feedback rule construction mechanism is used to issue feedback messages to learners in cases where the synchronous CMC support system detects that the learners have strayed astray from the expected learning topics in their posts. The classification rates for the one-class SVM classifier can reach up to 97.06%, and the average pre-test and post-test grades were 51.94 and 66.77, respectively, which revealed that the junior high school students participating in synchronous CMC activities related to natural science were benefited by the proposed intelligent synchronous CMC platform.