Environment Aspects and Daily Life-Threatening Risk Prediction for Improving Public Health Using Ensemble Learning Techniques

Abstract

Increasing work nature in the fastest world need relaxing to maintain public health is important. But the continuous imbalance of work nature in human daily life doesn’t have time to consecrate the health with leads life threatening facts such as work tension, management stress, job professional, family needs, improper periodic work cycle, and soon. By analyzing the daily public life threatening risk prediction based on Feature selection and classification to category the public health to recommend for phycological treatment.  By suggesting treatment schedule are important to balance the public health Lifecyle to make stress free life to spent with nature.  Most of prevailing techniques analyse the life-threatening issues, but the features are improperly to taken without the mutual relation cause poor accuracy in precision and classifications rate with more false prediction rate. To tackle this issue, to propose an optimized ensemble learning Techniques based on multi scalar support vector machine (MSSVM) with deep scaled XGboost classifier (DSXG-boost) to improve the prediction accuracy. Initially the public life Cycle dataset is collected and to make normalization using Min-max normalization. Then the Public Life Threat Impact Rate (PLTIR) is analysed to marginalize the heath affecting features. Then then stress margin dependencies feature limits are selected with support of MSSVM. The selected features are grouped in cluster margins and classified with deep scaled XG boost algorithm to predict the active and inactive heath margin be categorized by risk threaten class. The proposed system prove the prediction accuracy in higher precision are by selecting mutual dependencies of health affection feature limits and best recall rate to improve the performance. Based on the predicted class, the categorized life threaten risk peoples are recommends to make phycological treatment to protect the public life cycle.