Having all the information about learners and learning resources (metadata and reputation metadata), collaborative filtering was applied to recommend personalized learning resources.
The action part of an agent's rule can perform a number of recommendations or adaptations in a scenario: give advice to an actor, notify another actor than the one targeted by the rule, recommend various learning resources, update the user's model, and propose to jump to another activity or to another learning scenario.
When system servers receive real-time data from learner devices, intelligence and analytical tools such as learning analytics, data mining, and big data, are used to analyze and store the learning data, and then recommend learning patterns and resources to learners.
The Recommender System recommends topics and learning resources based on the student's past long-term performance and psychological profile.
For mid-career professionals who want to enter the field, Griffin recommends learning as much as possible about the industry via resources on the Internet and conversations with people in the field.
To gather data I recommend learning Google Analytics.
For beginners in program design, it is recommend learning "python".
At this point, we have made principled decisions on what learning resources to recommend to students at the right time.
In practical applications, this method can be integrated to customized recommendation systems that can recommend interesting hot topics associated with learning resources to cater for learners' individual learning demands.
It recommends: Creating a national online library of learning resources, with a central bank of tools, software and apps available to everyone.
Having faculty integrate these learning resources into their teaching, integrating interactive tools into this e-learning environment to foster interactions, and using mobile devices for convenient access are recommended to help students make more effective use of these resources.
