One improved learning analytics of interest transfer in interactive learning activities

Mining interactive learning activities and exploring learners' interest transfer are the key issues to realize education optimization and learning feedback. It also puts forward higher technical requirements. Based on the massive learning behaviors of online learning platform, this study proposes an interest transfer analysis method of interactive learning activities based on locality sensitive strategy. First, we analyze the multi-dimensional attributes and relationship characteristics; second, we design the improved locality sensitive hashing algorithm, then train multiple data sets of learning interactive activities. The experiments explore the performance indicators, mines the topological relationships of interest transfer, and evaluates and predicts the potential rules; finally, the interest transfer mechanisms supported by locality sensitive strategy are applied to the actual teaching process, and the locality sensitive activities tested in practice. It is of great practical significance to optimize the learning process, and it is also a technical reference for learning analytics to be deeply integrated into the interactive learning environments.