Home

In this project, we try to classify articles by their relevance to the field of software languages based on indicators. We learn indicator configuration from a small seed set and constantly improve it manually by comparing the results of individual indicators. For example, Infoboxes are reliable from the start, if they are available. The indicator based on the summary can then be improved by trying to recall every article that the infobox indicator would recall.