Stride Window Approach with Anomaly Detection for Probability Risk Assessment

Different data mining techniques are used for risk assessment and prediction purpose in different domains, such as finance sector, health sector, education sector, etc. The risk assessment is directly proportional to dependency of the factors concerning the task. In this paper, we proposed a novel generalized model for Probability Risk Assessment using Stride Window Approach for data augmentation with integrating the Anomaly Detection technique. This paper uses the Stride Window Approach to improve the data computation by reducing the bias and duplicate data, then applying the Anomaly Detection for risk assessment. The integration of the stride window with Anomaly Detection technique provides the double advantage in risk assessment process. The illustration of proposed model has been done on the financial datasets. Results show the importance of the proposed model for probability risk assessment based on the performance analysis of the temporal sequential data.