Named Entity Recognition for Smart City Data Streams: Enhancing Visualization and Interaction

EasyChair Preprint 15714

Abstract

The emergence of smart cities has generated an abundance of data from various urban sources, necessitating robust methods for processing this information efficiently. Named Entity Recognition (NER) is crucial for extracting meaningful insights from these data streams. In this research, we propose an innovative NER approach specifically designed for Smart City Data Streams, focusing on improving visualization and interaction for better analysis. Our method employs advanced deep learning techniques to accurately identify and classify entities from a range of sources, including traffic reports, social media feeds, and environmental sensors. By incorporating a multi-modal framework that utilizes contextual information and geographical metadata, we enhance the precision of entity recognition. The system enables real-time processing of urban data, allowing city planners and stakeholders to visualize the relationships between entities dynamically. To facilitate user interaction, we develop interactive dashboards and visualization tools based on NER outputs, supporting intuitive exploration of urban information. Evaluations conducted on real-world datasets reveal substantial advancements in entity recognition accuracy and efficiency relative to conventional techniques, significantly contributing to informed decision-making and increased public engagement in smart city projects.

Keyphrases: Named Entity Recognition, Smart City Data Streams, visualization

@booklet{EasyChair:15714,
  author    = {Xiang Li and Jeffrey Wang and Lihua Zhang},
  title     = {Named Entity Recognition for Smart City Data Streams: Enhancing Visualization and Interaction},
  howpublished = {EasyChair Preprint 15714},
  year      = {EasyChair, 2025}}