HUMAN MOBILITY PATTERN MINING: A SYSTEMATIC REVIEW OF METHODS AND DATA PROCESSING
HUMAN MOBILITY PATTERN MINING: A SYSTEMATIC REVIEW OF METHODS AND DATA PROCESSING

Summary/Abstract: Human mobility pattern mining has emerged as a significant research field, yet existing studies are relatively isolated and lack an integrated review of addressed issues and tested solutions. This systematic review aims to provide a comprehensive analysis of human mobility pattern mining research across three interconnected dimensions: data processing approaches, methodological landscape, and future research directions. Following PRISMA guidelines, this study systematically reviewed 43 carefully selected papers from Scopus-indexed journals, covering the period from 2018 to 2025. A structured two-phase approach was used to extract and categorize information according to four research questions, and then analyze patterns to generate insights and recommendations through comprehensive synthesis. The analysis revealed six distinct conceptual perspectives on human mobility research, five categories of real-world applications, and five common methodological approaches ranging from traditional statistical methods to advanced artificial intelligence techniques. Data quality assessment can be categorized into three fundamental dimensions: completeness, accuracy, and consistency. A four-phase preprocessing pipeline was developed with integrated quality control mechanisms. Current challenges include data quality limitations, temporal dimension inadequacies, and scalability barriers. This review provides a systematic organization of fragmented knowledge and identifies four key future research directions: enhanced data integration, advanced spatiotemporal modeling, semantic enhancement, and scalable computing infrastructure. These findings establish foundations for developing more robust, scalable, and practically applicable mobility analysis frameworks.

• Details
• Contents