Paper Title
Improving Education Performance for School Students Based on Life Skill Prediction and Development using an Optimized Machine Learning Decision Approach

Abstract
In recent day analysis, education is essential for every student seeded from school life to improve Skill set based on life skill education. To improve the student performance based on skill set analyses, prediction, and classification. Based on low-performance categories, students will be motivated to enhance their skill set by giving additional teaching and training to the student. Artificialintelligence development is essential in analyzing the student's skill set to make the prediction. But in some cases, the importance of the Skillset is appropriately evaluated because most machine learning models don't mean the importance of feature dependenciesrelated to skill set development. To resolve this problem, we propose a deep feature analysis using Hyper Spectral Support Vector Machine (HS2VM) based Artificial Neural Network (ANN) for predicting student performance. Initially, the student logs were collected from public and private school management to preprocess to scale the performance weights presence. Then Skillset interest score (S.I.S.) was evaluated to predict the features using support vector optimization. Further, the decision tree logic was applied to choose the importance of features. Finally, the selected features are trained with ANN to predict student performance and Rank to categorize classes. Based on the predicted type, the recommendations and developments were attained with new programs to motivate the students to improve their performance. This proposed system indicates the skillset feature analysis efficiently to perform better to produce higherprecision and recall rate to produce high classification accuracy. Keywords - Education and training, Machine learning, Children Performance Prediction, Life SkillSet Prediction, Feature Analysis, Prediction, and Classification.