Software Technology Immersive Education Research and Practice Based on Spark Platform

From the perspective of education informatization, a teaching resource sharing system with huge storage space and computation amount is established. When the user accesses the teaching materials, it constructs the database with a large number of users through instant access to the data. This project intends to study an optimization scheme that does not require any intermediary output based on Hadoop and combined with Spark's parallel computing mode, so that it can make better use of Spark's iterative computing capability. It realizes the efficient recommendation of teaching resources. A hierarchical resource model based on collaborative filtering is proposed. The unrated resources predicted in this way are imprecise. A time function is added to the prediction process for non-rated resources to reflect the change in user interest over time. The index function is used to describe the change of time. The practicability of the Spark-based in the recommendation of teaching resources are verified by the research.