NLP-Driven Proactive Risk Assessment of Road Construction

Abstract

Real-time and accurate extraction of road construction information is crucial for urban traffic management and power grid maintenance in smart city development. Traditional methods relying on manual reporting and government announcements are time-consuming and inefficient. This paper proposes a method integrating data mining and natural language processing (NLP) technologies to integrate multi-source, diverse road construction datasets for predicting risks associated with power grid lines. By utilizing advanced NLP techniques and pretrained models, key information such as construction sites, timings, and reasons are effectively extracted from the collected data, enabling efficient and predictive management of construction-related risks.