Effective recommendation in a knowledge base - the Skybrary case study
|has title::Effective recommendation in a knowledge base - the Skybrary case study|
|Master:||project within::Information Sciences|
|Student name:||student name::Roy Hoeymans|
|Supervisor:||Victor de Boer|
|Second supervisor:||Timo Kouwenhoven|
Skybrary is an internet platform launched by DNV-GL in 2008 that contains a large amount of aviation and air traffic management safety knowledge. Since its conception it has grown into a knowledge base that is internationally recognized and certified by the most important aviation organizations, institutions, and regulators. Each article contains a ‘Related Articles’ –section. This section contains links to articles that are related to it. Currently, this selection of links is compiled by the experts who also edit the articles. DNVGL wants to find out if it is possible to make this selection dynamically, by using a recommender system. The challenge of this research is to built a recommender system using a relatively small amount of data by experimenting with several types of data and algorithms.