A multilabel classification approach to identify hurricane‐induced infrastructure disruptions using social media data

Abstract

Rapid identification of infrastructure disruptions during a disaster plays an important role in restoration and recovery operations. Due to the limitations of using physical sensing technologies, such as the requirement to cover a large area in a short period of time, studies have investigated the potential of social sensing for damage/disruption assessment following a disaster. However, previous studies focused on identifying whether a social media post is damage related or not. Hence, advanced methods are needed to infer actual infrastructure disruptions and their locations from such data. In this paper, we present a multilabel classification approach to identify the cooccurrence of multiple types of infrastructure disruptions considering the sentiment toward a disruption-whether a post is reporting an actual disruption (negative), or a disruption in general (neutral), or not affected by a disruption (positive). In addition, we propose a dynamic mapping framework for visualizing infrastructure disruptions. We use a geo-parsing method that extracts location from the texts of a social media post. We test the proposed approach using Twitter data collected during hurricanes Irma and Michael. The proposed multilabel classification approach performs better than a baseline method (using simple keyword search and sentiment analysis). We also find that disruption-related tweets, based on specific keywords, do not necessarily indicate an actual disruption. Many tweets represent general conversations, concerns about a potential disruption, and positive emotion for not being affected by any disruption. In addition, a dynamic disruption map has potential in showing county and point/coordinate level disruptions. Identifying disruption types and their locations is vital for disaster recovery, response, and relief actions. By inferring the co-occurrence of multiple disruptions, the proposed approach may help coordinate among infrastructure service providers and disaster management organizations. 1 INTRODUCTION Cities and communities all over the world largely depend on critical infrastructure systems/services such as electrical power, water distribution, communication services, and transportation networks. The growing interconnectedness and © 2020 Computer-Aided Civil and Infrastructure Engineering interdependency among these systems have changed the organizational and operational factors and increased the vulnerability in the face of unwanted disruptions. These systems provide critical services to a large population, and thus when disrupted they affect our quality of life, local and regional economy, and the overall community wellbeing. The need to

A hybrid deep learning model for predictive flood warning and situation awareness using channel network sensors data

“…To do so, flood‐related variables such as flood duration, rainfall intensity, flow velocity, characteristics of buildings and infrastructures, as well as socioeconomic factors, are embedded in ML models to develop flood damage functions and flood loss function for predicting the extent of damage and loss in different flood scenarios (Kreibich, Botto, Merz, & Schröter, 2017; Merz, Kreibich, & Lall, 2013). ML techniques have also been widely utilized for flood hazard modeling such as inundation modeling (Alipour, Ahmadalipour, Abbaszadeh, & Moradkhani, 2020) and infrastructure disruption detection using crowdsourced data following hurricanes and floods (Roy, Hasan, & Mozumder, 2020).…”

Real-Time Twitter Data Mining Approach to Infer User Perception Toward Active Mobility

“…Existing text classification methods can be divided into a traditional machine learning approach and a deep learning approach. Naïve Bayes ( 11 ), logistic regression ( 12 , 13 ), and support vector machine ( 14 ) are the most commonly used machine learning approaches for text classification. Naïve Bayes is commonly used as a standard for text classification since it is quick and simple to execute ( 15 ).…”

“…The logistic regression-based multiclass text classification has shown superior performance compared with other traditional approaches ( 13 ). This algorithm assigns weights to each input sequence to segregate potential classes from each other ( 17 ).…”

Resilience assessment of electrified road networks subject to charging station failures

“…Social sensing is also being considered to help resilience management for infrastructures. Roy, Hasan, &Mozumder (2020), andC. Zhang, Yao, Yang, Huang, & used social media data to assess disruption types and locations of infrastructures (transportation systems, power systems, etc.)…”