Allometric scaling of road accidents using social media crowd-sourced data

Ghandour, Ali; Hammoud, Huda; Dimassi, Mohammad; Krayem, Houssam; Haydar, Jamal; Issa, Adam

doi:10.1016/j.physa.2019.123534

Cited by 6 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, in [13], a vehicle monitoring system (VMS) based on LoRa protocol is developed to monitor different environmental parameters, including ambient temperature and humidity, and air quality parameters, such as PM2.5, NO2, CO, and O3. In order to Collect Lebanon roads downtime events from social media, a real-time online Lebanese road accident platform is proposed in [14].…”

Section: Literature Reviewmentioning

confidence: 99%

Intelligent Bus Application for Smart City based on LoRa Technology and RBF Neural Network

Kanaan

Haydar

Samaha

et al. 2020

Wseas Transactions on Systems and Control

Self Cite

View full text Add to dashboard Cite

Nowadays smart transportation is one of the most important smart city applications for all the facilities it provides, as well, achieves green transportation. One of these applications include bus monitoring systems for schools and universities. This paper proposes an integrated bus tracking and monitoring system using Internet of Things (IoT) mainly LoRa technology for Islamic University of Lebanon (IUL) to ease the bus issue that comprises a barrier for good service. In the proposed system, we have implemented a case study to monitor the position, speed, humidity and temperature of IUL shuttle buses. In addition to the tracking, a prediction algorithm is implemented for computation of the arrival time of the bus to the Wardanieh campus. The estimation is based on artificial intelligence neural network mainly the Radial Basis Function (RBF) algorithms for the sake of finding the predicted time of arrival in different locations and scenarios starting from Khaldeh campus to the Wardanieh campus. The collected data are instantly shown on a user interface for monitoring, whereas the exact location and arrival time of the bus will be displayed on an android application for students. Results show that the arrival estimated time error is about 0.18%.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Intelligent Bus Application for Smart City based on LoRa Technology and RBF Neural Network

Kanaan

Haydar

Samaha

et al. 2020

Wseas Transactions on Systems and Control

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several previous studies have used social media data (Ghandour et al 2020;Ghandour, Hammoud, and Telesca 2019;Gu, Qian, and Chen 2016;Kutela, Magehema, et al 2022;Zhang et al 2018). For instance, Zhang et al (2018) applied deep learning to detect traffic accidents from social media data in two metropolitan areas: Northern Virginia and New York City.…”

Section: Introductionmentioning

confidence: 99%

Leveraging Social Media Data to Understand Spatial and Severity of Roadway Crashes in Tanzania

Kutela

Mwekh’iga²,

Kilaini³

et al. 2022

JSS

View full text Add to dashboard Cite

Traffic crashes are among the major cause of fatalities in most developing countries. A lack of a centralized historical crash database impedes the efforts to perform large-scale studies to understand the causes of crashes. However, the information for most major crashes are normally shared on social media. Therefore, this study used social media posts to extract information related to head-on and run-off roadway (ROR) crashes that occurred between 2010 and 2020. A total of 661 crashes were collected, which included 409 head-on and 252 ROR crashes. Geographical Information System (GIS) applications and multinomial logit models were developed to understand the spatial distribution of crashes and the resulting severity. It was found that the spatial distribution of crashes differs significantly by type and severity. Head-on collisions were predominant in two eastern regions, while ROR collisions appeared to spread on a large scale across the entire country. The pattern of fatal collision includes one southern region in addition to the two Eastern regions. Conversely, the pattern of non-injury crashes shows predominance along the coastal area. The Multinomial model results showed that speeding was two times more likely to result in fatalities for ROR crashes. Crashes in rural areas were about two times more likely to be fatal, given that they were head-on collisions. Further, bus-involved crashes were less likely to result in injuries. The implications of these study findings to the practitioners are discussed.

show abstract

“…The last group is oriented to the amount of data collected by all the technology currently available in the community. Among them can be found data from social networks [12,17,50,51], data collected from sensors of a smartphone [13], structured data [52,53], data detected from video cameras [14,54,55], and traffic sensor data [56]. These datasets are widely used by deep learning, hybrid, and extreme learning methods.…”

Section: Introductionmentioning

confidence: 99%

Automatic Detection of Traffic Accidents from Video Using Deep Learning Techniques

2021

View full text Add to dashboard Cite

According to worldwide statistics, traffic accidents are the cause of a high percentage of violent deaths. The time taken to send the medical response to the accident site is largely affected by the human factor and correlates with survival probability. Due to this and the wide use of video surveillance and intelligent traffic systems, an automated traffic accident detection approach becomes desirable for computer vision researchers. Nowadays, Deep Learning (DL)-based approaches have shown high performance in computer vision tasks that involve a complex features relationship. Therefore, this work develops an automated DL-based method capable of detecting traffic accidents on video. The proposed method assumes that traffic accident events are described by visual features occurring through a temporal way. Therefore, a visual features extraction phase, followed by a temporary pattern identification, compose the model architecture. The visual and temporal features are learned in the training phase through convolution and recurrent layers using built-from-scratch and public datasets. An accuracy of 98% is achieved in the detection of accidents in public traffic accident datasets, showing a high capacity in detection independent of the road structure.

show abstract

Allometric scaling of road accidents using social media crowd-sourced data

Cited by 6 publications

References 14 publications

Intelligent Bus Application for Smart City based on LoRa Technology and RBF Neural Network

Intelligent Bus Application for Smart City based on LoRa Technology and RBF Neural Network

Leveraging Social Media Data to Understand Spatial and Severity of Roadway Crashes in Tanzania

Automatic Detection of Traffic Accidents from Video Using Deep Learning Techniques

Contact Info

Product

Resources

About