Recently, the widespread diffusion of ICT technologies enabling the Internet of Things has transformed many cities into “smart cities” motivated by the increasing citizens' need of a more comfortable and livable ambient. In recent years various works proposed innovative solutions to help administrations in monitoring various aspects of the city life and the connected services (e.g, the air and noise pollution monitoring to enhance public transportation), but these solutions usually base their proposals only on one of the two most spread approaches: the monitoring based on the data collected through IoT sensors, and the one based only on the suggestions received by users about dangerous or strange situations in the area of the city in which they live. For this reason, this paper proposes an innovative architecture to combine the two approaches. A modular architecture is proposed to support a) the collection of citizens' feedback (positive or negative) in terms of free-text messages, b) the automatic categorization (positive / negative) of received free-text comments performed through both a consolidated topic-detection algorithm and a Machine Learning algorithm based on the sentiment analysis approach, c) the integration of mobile devices and fixed devices for monitoring different areas of the city with respect to the services reviewed by the citizens, d) the visualization of the status of the city and its areas in a dedicated dashboard capable of correlating the two types of data and providing statistics on the real conditions of the areas. A preliminary prototype of the whole architecture and the described mobile and fixed devices was developed and tested to demonstrate the feasibility of the approach and inspire new future works.
An Innovative Decision Support System for Smart Cities Government based on Sentiment Analysis and IoT technologies
Teodoro Montanaro;Ilaria Sergi;Enza Giangreco;Luigi Patrono
2022-01-01
Abstract
Recently, the widespread diffusion of ICT technologies enabling the Internet of Things has transformed many cities into “smart cities” motivated by the increasing citizens' need of a more comfortable and livable ambient. In recent years various works proposed innovative solutions to help administrations in monitoring various aspects of the city life and the connected services (e.g, the air and noise pollution monitoring to enhance public transportation), but these solutions usually base their proposals only on one of the two most spread approaches: the monitoring based on the data collected through IoT sensors, and the one based only on the suggestions received by users about dangerous or strange situations in the area of the city in which they live. For this reason, this paper proposes an innovative architecture to combine the two approaches. A modular architecture is proposed to support a) the collection of citizens' feedback (positive or negative) in terms of free-text messages, b) the automatic categorization (positive / negative) of received free-text comments performed through both a consolidated topic-detection algorithm and a Machine Learning algorithm based on the sentiment analysis approach, c) the integration of mobile devices and fixed devices for monitoring different areas of the city with respect to the services reviewed by the citizens, d) the visualization of the status of the city and its areas in a dedicated dashboard capable of correlating the two types of data and providing statistics on the real conditions of the areas. A preliminary prototype of the whole architecture and the described mobile and fixed devices was developed and tested to demonstrate the feasibility of the approach and inspire new future works.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
