Analysis of Fire Accident Factors on Construction Sites Using Web Crawling and Deep Learning Approach

Kim, Jae Hong; Youm, Sangpil; Shan, Yongwei; Kim, Jong-Hoon

doi:10.3390/su132111694

Cited by 10 publications

(6 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The crawler receives a prioritized set of URLs (Uniform Resource Locators) to visit and collects hyperlinks from those pages. It iteratively downloads the content of the web pages until the specified or predefined criteria are met, such as a certain number of pages visited, a particular keyword or phrase found, or a specific amount of time elapsed (Moon et al, 2018;Cho, 2001;Olston and Najork, 2010).…”

Section: Web Crawlermentioning

confidence: 99%

“…The Internet, a globally connected network system that enables worldwide access, provides a wide range of information resources through a diverse collection of private, public, business, academic, and government networks. These resources, including news, articles, research papers, interviews, reports, and official documents, are valuable for gathering country information for an effective risk assessment (Moon et al, 2018;Shin, 2015;Pathirage et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

“…The rapid expansion/growth of the Internet has contributed to the accumulation of country data, which is continuing to increase (Shin, 2015;Hjelt and Björk, 2006). However, acquiring timely, precise, and accurate information about host countries can be challenging and time-consuming, requiring qualitative and quantitative assessments that depend on numerous factors and may involve mental shortcuts, heuristics, or biases (Moon et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Using Web Crawling for Automated Country Risk Assessment

Ozyurt,

Tah,

Dikmen

et al. 2023

Proceeding of the 33rd European Safety and Reliability Conference

View full text Add to dashboard Cite

The latest and most accurate host country information is crucial for international market selection and bidding decisions. In today's digital world, this information can be timely gathered from websites on the Internet, the globally connected network system facilitating worldwide access to a wide range of information resources through a massive collection of private, public, academic, and government networks. Within construction companies, the traditional risk assessment process is generally based on searching for country information on the web and necessitates human input, which is subjective by nature and prone to misinterpretation and errors. With the help of automation, the traditional way of identifying and assessing country risks can change for the better. Better, latest, and most accurate data collection with web crawling algorithms, which are used for automated browsing, may be possible for country risk assessment with less human involvement, minimizing human errors and saving effort and time. Studies that explored the automated collection of information from the Internet for country risk assessment in construction are limited in the literature. In this paper, we assert that the web crawling technique can help search for data from the web and can be used to automatically develop country risk registers. Within this context, first of all, the web crawling algorithms will be explained, and how they can be used to automate the risk assessment process will be discussed. Then, a demonstrative example of an international construction company carrying out a country risk assessment process while preparing a bid for a highway project in Serbia will be depicted. The benefits of web crawling for automated country risk assessment will be discussed as well as its challenges.

show abstract

Section: Web Crawlermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Using Web Crawling for Automated Country Risk Assessment

Ozyurt,

Tah,

Dikmen

et al. 2023

Proceeding of the 33rd European Safety and Reliability Conference

View full text Add to dashboard Cite

show abstract

“…It is suggested that most of the causes for absence are directly related to the activities of the workers and their recommended way of performing them [10,11]. However, the adoption of complex ergonomic solutions at construction sites requires long times and the involvement of a wide range of different stakeholders [12].…”

Section: Introductionmentioning

confidence: 99%

“…In this context, the rise of the Internet of Things (IoT) devices and wearable technologies provides for a relatively unobtrusive technology that could enhance safety by helping in the monitoring of the workers' activities and their adherence to the recommended ergonomic and risk-avoidance guidelines. In addition, the analysis of abnormal ways of performing activities, such as walking very slowly or around objects, could provide insight on the conditions of the construction site, allowing taking important measures that mitigate safety risks [11].…”

Section: Introductionmentioning

confidence: 99%

Introducing VTT-ConIot: A Realistic Dataset for Activity Recognition of Construction Workers Using IMU Devices

et al. 2021

View full text Add to dashboard Cite

Sustainable work aims at improving working conditions to allow workers to effectively extend their working life. In this context, occupational safety and well-being are major concerns, especially in labor-intensive fields, such as construction-related work. Internet of Things and wearable sensors provide for unobtrusive technology that could enhance safety using human activity recognition techniques, and has the potential of improving work conditions and health. However, the research community lacks commonly used standard datasets that provide for realistic and variating activities from multiple users. In this article, our contributions are threefold. First, we present VTT-ConIoT, a new publicly available dataset for the evaluation of HAR from inertial sensors in professional construction settings. The dataset, which contains data from 13 users and 16 different activities, is collected from three different wearable sensor locations.Second, we provide a benchmark baseline for human activity recognition that shows a classification accuracy of up to 89% for a six class setup and up to 78% for a sixteen class more granular one. Finally, we show an analysis of the representativity and usefulness of the dataset by comparing it with data collected in a pilot study made in a real construction environment with real workers.

show abstract