Intelligent pedagogical agents (IPA) are aimed to support learning in virtual worlds. Motivations for adopting IPAs in virtual worlds are to compensate for lack of human pedagogical presence, to improve student engagement, and having autonomous support. Given named challenges to realizing IPAs in virtual worlds, a proposed solution approach is to simulate IPAs with targeted scenarios with intel...

To boost the application of machine learning (ML) techniques for credit scoring models, blackbox problem should be addressed. The primary aim this paper is to propose a measure based on counterfactuals evaluate interpretability ML technique. Counterfactuals assist with understanding model regard classification decision boundaries and robustness. second contribution development data perturbation...

New learning methods are emerging in order to improve and facilitate the use of communication and informatics technologies both by students and professors. This paper describes an Ambient Intelligence application scenario based on Computer Supported Collaborative Learning and MANET in order to create creative and flexible e-learning environments where students can use wireless technologies for ...

Collaborative learning environments are mainly based on constructivist instructional design theories where students construct their own knowledge by interacting with their environment. The so-called virtual communities constitute and approach that facilitates the constructivist learning within these environments. In this context, the learning process consists of authentic participation in the a...

Individualized learning is the most important process that need to use individualization constraints to ensure adaptive systems able to support flexible solutions dynamically adapt content as well as interface and the scenario of multimedia learning object to fit pedagogical intentions. Therefore, this various aspect of multimedia learning object have to be supported in a highly personalized ma...

background: video-based instruction has been stated as an effective method of teaching and its potentials have encouraged instructors learners to opt this method purpose: the current study has compared the students' cognitive learning in the two approaches of lecture- based and video-based teaching. methods: this quasi-experimental study was conducted on 88 senior nursing students who were rand...

Active learning methods seek to reduce the number of labeled examples needed to train an effective classifier, and have natural appeal in spam filtering applications where trustworthy labels for messages may be costly to acquire. Past investigations of active learning in spam filtering have focused on the pool-based scenario, where there is assumed to be a large, unlabeled data set and the goal...