Efficient Intelligent Generic Recommendation Knowledge Graph in Education Domain using Association Rule Mining and Machine Learning

Abstract

Knowledge graph is used to extract and derive new facts from huge variety of data sources through relationship. An existing Natural language processing tool is specific and performs adaptive learning mechanism through instruction concepts. The specific knowledge graph suffers a problem of finding large collections of new facts with inter domain. This problem is addressed by implementing an efficient model for integrating various domain of interest as a generic knowledge graph. This proposed model has three major phases they are generic data collection, generic relationship establishment and generic deployment for education domain. The data are collected, preprocessed and categorized in to specific subject category by producing integrated data set. The relationship is established based on the pedagogical data with assessment data of leaners are classified in to course list. This generic knowledge graph is compared with the CNN based model and GCN based model. The validation of these models are assessed and deployed into application services for teachers and learners. The main objective of the proposed graph is to organize a generic knowledge graph for deriving huge amount of new facts to the education domain with maximum support and confidence level.