Machine Learning Analytic-Based Two-Staged Data Management Framework for Internet of Things.

Area of Science:

• Computer Science
• Data Science
• Network Engineering

Background:

• Internet of Things (IoT) applications involve numerous connected nodes with strict resource constraints (battery, processing, storage).
• Standard data management methods are insufficient for these highly constrained and numerous IoT environments.
• Machine learning offers a promising approach to address these complex management challenges.

Purpose of the Study:

• To design and implement a novel framework for efficient data management in IoT applications.
• To address the limitations of existing methods in managing resource-constrained IoT networks.
• To improve the overall performance and longevity of IoT devices.

Main Methods:

• Developed the Machine Learning Analytics-based Data Classification Framework (MLADCF).
• MLADCF employs a two-stage approach combining a regression model and a Hybrid Resource Constrained KNN (HRCKNN).
• The framework learns from real-world IoT application analytics for adaptive management.

Main Results:

• MLADCF demonstrated proven efficiency across four diverse datasets compared to existing approaches.
• The framework significantly reduced global network energy consumption.
• Extended battery life for connected IoT nodes was observed.

Conclusions:

• MLADCF provides an effective solution for data management in resource-constrained IoT environments.
• The proposed framework enhances the efficiency and sustainability of IoT networks.
• Machine learning integration is crucial for optimizing future IoT data management strategies.