Data-driven quality assurance in career and lifelong education supported by Individual Learning Accounts

Quality assurance (QA) in lifelong and career education guarantees that learning experiences are successful and relevant. However, traditional QA approaches have drawbacks, including fragmented data, manual evaluation, delayed feedback, and inadequate scalability. The research presents an AI-driven quality assurance approach that uses extensive learner data from IndividualLearning Accounts (ILA) to provide real-time, personalized quality monitoring and recommendations. It proposes a Dynamic Shark Smell optimized Convolutional Support Vector Machine (DSS-CSVM) to improve the accuracy and efficiency of quality assurance in career and lifelong education by optimizing model parameters for precise course effectiveness prediction and anomaly detection. The dataset comprises extensive learner records obtained from ILA, including course completion data, assessment scores, and learner feedback from multiple online education platforms. Z-score normalization is used to standardize numerical characteristics. The system utilizes the DSS method to optimize the parameters of a CSVM, which predicts course efficacy and identifies abnormalities in learning outcomes. It offers educators and learners actionable information through the collection of detailed learning histories, assessment data, and feedback fromILA. The proposed DSS-CSVM model predicts course quality and detects anomalies with an accuracy of over 96%. The proposed DSS-CSVM model achieved a precision of 0.941, a recall of 0.921, an F1-score of 0.936, and an accuracy of 96.28%, demonstrating its effectiveness for accurate and reliable quality assurance in lifelong and career education. This AI framework, based on the proposed suggestion, improves QA for lifelong education by giving scalable and customized solutions using DSS-CSVM.