Enhancing Trajectory Data Privacy through Information Gain-Fused Semantic Generalization

This research proposes the Information Gain-Fused Semantic Generalization Trajectory (IGFSGT) model to address the challenges of privacy preservation in digital data. Traditional trajectory data often suffers from inaccuracies and variations, hindering its effectiveness in classification and generalization. The IGFSGT model employs an optimized information gain utility function to achieve accurate node segmentation while adhering to privacy requirements. By adjusting threshold values, the model ensures that the most informative splits are selected, maximizing information gain. To further enhance performance, the model incorporates nodal transformation through semantic distances and fuzzy logics. Experimental results demonstrate the superior performance of IGFSGT in terms of both privacy protection and data utility compared to existing models. This research contributes to the development of privacy-preserving data publication techniques, enabling the sharing of valuable information while safeguarding sensitive data.