A survey of hierarchical classification algorithms with big-bang approach

Hierarchical classification is a technique used to solve problems with hierarchical concepts, which are usually arranged into trees or directed acyclic graphs (DAGs). Research in this field is mostly done in the field of bioinformatics and text classification, because both fields have hierarchical problems. Completion of hierarchy classification can be done with a local approach and a big-bang approach. From several previous studies, it was found that hierarchical classification with the big bang approach got good results, both in terms of predictive accuracy, model size and time needed to build the model. In this paper, we survey previous research on hierarchical classification algorithms using bing-bang approach.