Smart education system to improve the learning system with CBR based recommendation system using IoT

Over the last few years, the research fields of intelligent learning systems have been improving the process of learning systems. Smart Tutoring System-(STS) applications have been used in e-learning. The results signify the importance of the learner's engagement in customizing a model. The design outcomes of this IoT-based personalized learning system purely work on the audience's learning requirements, their keyword search, their learning experience levels, proficiency level of subjects, and the type of the course being taught. Students spent an average of 25.67 h accessing textual materials weekly, 27.4 h accessing video assets weekly, and similarly 6.5 h accessing visual learning materials. The research analysis part concludes participants' percentage of learning as per evaluations assessment, which increases whenever the outcome analysis comes with Case-Based Reasoning classifiers-CBR based search model. The findings displayed significant differences before and after learning case by case for every learner as per chosen topic and quiz assessments: 42.57% of the students responded before learning the first question assessments whereas 74.82%, of the students responded, after completion of learning from online resources based on their choice with CBR. Recommendation Model Analysis discussed root means square error-RMSE lies from 10% to 20% for 550 students group size. The RMSE result is 24% for a size of 1600, which is low performance compared to other group sizes. This study focuses on the STS recommendation model for the slow learner group to identify required learning from various online resources.