RTU Research Information System
Latviešu English

Publikācija: Modeling of Collaboration among E-Learning User Communities and Tagging Services

Publication Type Full-text conference paper published in other conference proceedings
Funding for basic activity Unknown
Defending: ,
Publication language English (en)
Title in original language Modeling of Collaboration among E-Learning User Communities and Tagging Services
Field of research 2. Engineering and technology
Sub-field of research 2.2 Electrical engineering, Electronic engineering, Information and communication engineering
Authors Daiga Balode
Sarma Cakula
Keywords collaborative tagging, knowledge item, tagging algorithm, tagging service, SVM - support vector machines; faceted browse
Abstract E-learning resources are developed by people having very different agendas and they need common meta-information framework to enable finding them across organizations. This paper focuses on collaboration among servers containing E-learning resources in different educational institutions for providing sharing and reusing of E-learning materials. Research problem is how to investigate collaborating servers, called tagging services, as a self-evolving infrastructure, which would allow sharing and reusing the Elearning materials in different user communities. The goal of the paper is to model a cluster of tagging services, storing E-learning materials or knowledge items such as bookmarks or index-cards and promote sharing and reusing of them. This paper considers a case when each service is situated separately – in different educational institutions, so each service is used by certain user community. E-learning materials put in system are marked with appropriate tags. Tags are taken from either an open or a closed vocabulary. In order to improve user experience and consistency, the tagging service offers a user a list of suggested tags, and to improve quality of suggestions, it also broadcasts request to other similar tagging services. The research idea is to use an algorithm based on self-learning SVM (Support Vector Machine) and graph theory for tagging service collaboration. Users should retain freedom to classify E-learning materials as they see fit, but they may benefit of being nudged in the right direction, e.g. given prompts about possible annotations and warned about mistakes or misspellings. Model has been tested short range in real environment.
Reference Balode, D., Cakula, S. Modeling of Collaboration among E-Learning User Communities and Tagging Services. In: 2nd WSEAS/IASME International Conference on Energy Planning, Energy Saving, Environmental Education, Greece, Corfu, 26-28 October, 2009. Corfu: WSEAS/IASME, 2009, pp.59-65.
ID 6863