Designing Frameworks for Integrating Spatio-Temporal Correlation for Efficient Pattern Extraction

AIM: The study seeks to compare the effectiveness of the DFS-MINE algorithm with that of using spatiotemporal correlations to analyse and enhance trip time prediction effectiveness. The general objective of this project is to apply real-time traffic data in improved prediction accuracy and effectiveness. Materials and Procedures: The two groups under study are labelled as follows. The 250 sample Integrating Spatio-temporal Correlations model has been labelled Group 1 while the 200 sample DFS-MINE model has been labelled Group 2. The cut-off is 0.05% confidence level 95% and the G Power value is 80%.Result: The Integrating Spatio-temporal Correlations model is much more efficient and predictive than the DFS-MINE approach. The prediction accuracy of DFS-MINE ranged from 72% to 81%, whereas the prediction accuracy of integrating spatiotemporal correlations ranged from 85% to 94%. Output frequency of the Integrating Spatio-temporal Correlations model for optimal efficiency has a significance level of around 0.00065. Conclusion: The study shows that the Integrating Spatio-temporal Correlations model is a superior method for optimizing trip time predictions since it is far more efficient than the DFS-MINE algorithm.