Adaptive recommendation for question decomposition in Web-based investigative learning

The Web allows learners to search for a huge number of Web resources/pages to investigate any question in a self-directed way. In Web-based investigative learning, learners need to decompose a question into sub-questions to create a learning scenario for elaborate investigation. In our previous work, we have developed a cognitive tool called interactive Learning Scenario Builder (iLSB). However, we have found that some learners finish investigative learning with insufficient question decomposition even if they use iLSB. This suggests the necessity of promoting question decomposition, which requires iLSB to recommend sub-questions to be decomposed in an adaptive way. This paper proposes a method for diagnosing insufficiency of learning scenario created by learners to extract and recommend keywords representing sub-questions to be decomposed by means of Linked Open Data (LOD), which have some relevance to questions decomposed within the scenario. This paper also reports a case study whose main purpose was to ascertain whether such an adaptive recommendation could promote wider and deeper question decomposition. The results suggest that it contributes to promoting learning scenario creation.