Next Place Prediction Based on Spatiotemporal Pattern Mining of Mobile Device Logs

Area of Science:

• Computer Science
• Mobile Computing
• Data Mining

Background:

• The proliferation of mobile devices has led to a surge in location-aware services.
• Accurate prediction of user next places is essential for proactive information delivery.

Purpose of the Study:

• To introduce a data-driven framework for predicting users' next places.
• To leverage past visiting patterns from mobile device logs for enhanced prediction.

Main Methods:

• Proposed the concept of spatiotemporal-periodic (STP) patterns to capture location visit periodicity.
• Developed algorithms to extract STP patterns from user mobility data.
• Utilized real-world mobile device logs for analysis and prediction.

Main Results:

• The proposed framework effectively identifies spatiotemporal-periodic patterns in user movements.
• Experimental results demonstrate superior performance compared to existing next place prediction methods.
• The approach shows high accuracy in predicting individual users' future locations.

Conclusions:

• The data-driven framework using STP patterns offers a significant advancement in next place prediction.
• This method enhances the capabilities of proactive information services in mobile environments.
• The findings validate the effectiveness of analyzing spatiotemporal periodicity for user behavior prediction.