A New Approach to Predict user Mobility Using Semantic Analysis and Machine Learning.

Area of Science:

• Computer Science
• Telecommunications Engineering

Background:

• Mobility prediction is crucial for seamless network handovers and location-based services.
• Existing methods often rely on extensive user mobility history, limiting their applicability.

Purpose of the Study:

• To develop a framework for user mobility prediction with and without historical data.
• To evaluate the effectiveness of using Short Message Service (SMS) and geological coordinates for mobility prediction in data-scarce scenarios.

Main Methods:

• Utilized Naïve Bayesian classification and Markov Models for predictions with mobility history.
• Employed Short Message Service (SMS) data and instantaneous geological coordinates for predictions in the absence of mobility history.
• Compared proposed techniques against the standard Markov Chain model.

Main Results:

• The proposed framework demonstrates superior performance by integrating spatial and temporal information.
• Accurate user future location prediction was achieved even without prior mobility patterns.
• The method shows significant promise for real-world applications like predicting medical rescue vehicle movements.

Conclusions:

• The developed framework offers a robust solution for mobility prediction, adaptable to varying data availability.
• Integrating diverse data sources like SMS and geological coordinates enhances prediction accuracy.
• The approach has practical implications for optimizing logistics and emergency response systems.