A Web-Based Recommendation System for Engineering Education E-Learning Solutions

The e-learning market consists of a wide variety of products, and it is still growing. To find an e-learning solution which fits the particular and situational demands is a very time consuming task, especially for teachers. Moreover, the technical and operational differences between the e-learning solutions are often not easy to understand from the product data and thereby consequences of choices are maybe not clear to the teacher. To solve these problems, a web-based recommendation system for teachers of engineering education is under development. This system is planned to support the decision making process of teachers about the use of an e-learning system. The precondition for setting up a recommendation system is that the desired entries (e.g. products, solutions, music or movies, etc.) are comparable to allow the algorithm to recommend: The current approach is to develop an e-learning scheme and compare the solutions based on this scheme. The determining of the necessary information for each e-learning solution has to be at least a half-automated process to keep the information up-to-date. Since expensive human time is needed to handle the necessary information, some approach with data and text mining as well as text analytics is promising. After the determining phase, each e-learning solution is represented by its data sheet. Apart from the e-learning solutions, also a teacher’s requirements have to be comparable to e-learning solutions to allow the algorithm to recommend. That is why a web-based questionnaire is utilized to catch the teachers’ requirements. A visual user-flow programming language is under development to provide an adequate environment for the development of the questionnaire and as an interface between the e-learning scheme, questionnaire and user-flow. The next step is to develop a functional prototype for the essential text analysis process to proof the concept. It is also required to analyze the current state of the e-learning scheme further to identify clusters of similar subjects, and to identify all critical properties to provide individual solutions for these.