INTEGRATING METHODS TO BUILD A RECOMMENDER SYSTEM WITH ADAPTED HELP
Universidad de Costa Rica (COSTA RICA)
About this paper:
Appears in: EDULEARN11 Proceedings
Publication year: 2011
Conference name: 3rd International Conference on Education and New Learning Technologies
Dates: 4-6 July, 2011
Location: Barcelona, Spain
Abstract:The purpose of this research is to identify how a Web-based authoring tool for the automatic generation, test and recommendation of examples in an Educational Adaptive Hypermedia (EAH) can be created.
We are using a recommender that reduces the complexity of finding an appropriate example for a particular teacher. Our system uses a probabilistic recommender supported by a Bayesian classifier. The recommendation in ARIALE is based on the automatic generation of examples, on the reuse of existing examples in a case base, and on the learning of the teacher’s decisions related to which examples to use. ARIALE keeps, classifies and use data related to the examples used by each teacher.
This paper explains the general structure and the implementation of our recommender system ARIALE (Authoring Resources for Implementing Adaptive Learning Environments). We describe a Web site, the three-tier architecture of the system, its knowledge-base and the modules (Session Manager, Planner and Helper) that process that knowledge to make decisions such as building a Web page and recommending examples adapted to a particular teacher. The sections in our paper also explain the layout of different Web pages and the general functionality that the system performs.
We also detail the method used for building our system, principally the aspects related to the design of the Web-site, the authoring tool, with emphasis on tan editor of network topologies. Because we did not find a complete method to develop intelligent help systems for authoring tools, we had to create a method for building our hypermedia system. Our method integrates fragments from other methods and guides for building Intelligent Tutoring Systems, Educational Adaptive Hypermedia, help systems, and adaptive interfaces.
This is a portrait of the complexity of building an educational adaptive hypermedia. The importance of this portrait is that it shows the integration of many methods and procedures that are usually studied disconnectedly. This integration results in a methodology to build an intelligent help system and its context in two steps: 1-the construction of the Web-site with the main application or functionality (the authoring tools), and 2- the intelligent help. This integration is needed because the authoring tool and the help must be coordinated.
The system is a “decision-maker” that uses knowledge to guide the construction of Web pages by selecting low-level resources such as a string of text or an image file.
Just like writing, painting or making a movie, the creation of multimedia applications or Web sites involves tools, media, and basic rules to structure messages to facilitate the understanding, acceptance and use by the people who consume these messages. The organization of events, widgets, and help contents compounds a particular syntax to provide the meaning of a concept or components of a Web page according to the current teacher’s context. This method of interaction explained in this paper becomes a grammar that can be reused by help systems designers.
Finally, we discuss the results of an evaluation of ARIALE.
Keywords: Recommender System, Machine Learning, Adaptive Help, Methods for building Adaptive Educational Hypermedia, Problem-solving support.