Intelligent Transportation Activity Recognition Using Deep Belief Network

Abstract

Comprehending and analyzing a diverse range of transportation modalities within urban environments is paramount for efficient traffic management and the development of smart cities. This paper explores a novel methodology for Transportation Activity Recognition (TAR) using data derived from GPS sensors, highlighting the potential to discern and categorize distinct modalities such as walking, cycling, and vehicular transport. Utilizing machine learning and advanced feature extraction for GPS sensors, the research processes and analyzes the GPS Microsoft Geo-life dataset, aiming to accurately identify and differentiate between diverse transportation activities and patterns. These activities' effects on urban traffic flow, congestion, and transportation planning are investigated, yielding insightful information that can improve and inform traffic management plans and regulations. The findings emphasize the potential of employing GPS sensors for a detailed and activity-specific analysis of transportation modes, contributing significantly to the evolution and implementation of intelligent and sustainable urban transportation systems.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 28 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.