Skip navigation
User training | Reference and search service

Library catalog

Content aggregators
Please use this identifier to cite or link to this item:

Title: Data-driven approach for incident management in a smart city
Authors: Elvas, L. B.
Marreiros, C. F.
Dinis, J. M.
Pereira, M. C.
Martins, A. L.
Ferreira, J. C.
Keywords: Buildings
Incident management
Knowledge extraction
Smart cities
Critical infrastructures
Issue Date: 2020
Publisher: MDPI
Abstract: Buildings in Lisbon are often the victim of several types of events (such as accidents, fires, collapses, etc.). This study aims to apply a data-driven approach towards knowledge extraction from past incident data, nowadays available in the context of a Smart City. We apply a Cross Industry Standard Process for Data Mining (CRISP-DM) approach to perform incident management of the city of Lisbon. From this data-driven process, a descriptive and predictive analysis of an events dataset provided by the Lisbon Municipality was possible, together with other data obtained from the public domain, such as the temperature and humidity on the day of the events. The dataset provided contains events from 2011 to 2018 for the municipality of Lisbon. This data mining approach over past data identified patterns that provide useful knowledge for city incident managers. Additionally, the forecasts can be used for better city planning, and data correlations of variables can provide information about the most important variables towards those incidents. This approach is fundamental in the context of smart cities, where sensors and data can be used to improve citizens’ quality of life. Smart Cities allow the collecting of data from different systems, and for the case of disruptive events, these data allow us to understand them and their cascading effects better.
Peer reviewed: yes
DOI: 10.3390/app10228281
ISSN: 2076-3417
Accession number: WOS:000594266500001
Appears in Collections:BRU-RI - Artigos em revistas científicas internacionais com arbitragem científica
ISTAR-RI - Artigos em revistas científicas internacionais com arbitragem científica

Files in This Item:
File Description SizeFormat 
applsci-10-08281.pdfVersão Editora5.06 MBAdobe PDFView/Open

FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Currículo DeGóis 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.