A Novel Approach Based on Fuzzy Rule and LSOWL–CNN Forecasting Students with Dropout Prediction and Recommendation Model

The main goal of governments is to guarantee that every individual worldwide, regardless of disability, have access to education. However, learners with disabilities exhibit higher rates of school and college dropouts than regular students. The main goal of contemporary research is to anticipate such drop-outs slightly earlier and offer recommendations and alternative pathways for their career enhancement by utilizing a novel approach LSOWL–CNN method. Initially, the dataset is pre-processed to increase the classification performance. Following that, the features are extracted and by employing ENT-QDA, the features are reduced. Rules based on fuzzy data will be created based on the features. A classification method called Linear Scaling Owl Optimal control Method with CNN (LSOWL–CNN) is employed to train the system using the created rules and testing the data the final results are classified. Results of the experiments showed that the suggested model attained an Accuracy of 96.817% which outperforms the existing models which is implementing in a python tool.