Dependency Extraction from Growth Trajectory using Sequential Pattern
Shunsuke Aoki, Ryosuke Saga, Takeo Ichinotsubo, Woyuan Niu, Hiroshi Tsuji · 2013
To extract learning order dependencies, we propose a method for analyzing growth trajectories. Due to limitations of the ISM-based method, weaker dependencies considered negligible for most learners' growth are likely to be lost. The proposed method, which is based on sequential pattern mining and minimum support, is expected to extract such dependencies. In addition, for complex dependencies, we can uniquely determine learning order by applying a pruning method based on supports. A case study of offshore software development is also discussed to verify the applicability of the proposed method.