Intersense: An XGBoost model for traffic regulator identification at intersections through crowdsourced GPS data

Vlachogiannis, Dimitris M.; Moura, Scott J.; Macfarlane, Jane

doi:10.1016/j.trc.2023.104112

Cited by 13 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fig 7 displays the confusion matrix, which shows that only eight fatal collisions were misjudged, compared to 167 correct predictions based on the test data. XGBoost can avoid overfitting using sophisticated optimization and regularisation methods such as weighted quantile sketching [ 31 , 32 ]. This helps to strengthen the generalizability of the neural network.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Comparing fatal crash risk factors by age and crash type by using machine learning techniques

Alshehri,

Alanazi,

Yosri

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

This study aims to use machine learning methods to examine the causative factors of significant crashes, focusing on accident type and driver’s age. In this study, a wide-ranging data set from Jeddah city is employed to look into various factors, such as whether the driver was male or female, where the vehicle was situated, the prevailing weather conditions, and the efficiency of four machine learning algorithms, specifically XGBoost, Catboost, LightGBM and RandomForest. The results show that the XGBoost Model (accuracy of 95.4%), the CatBoost model (94% accuracy), and the LightGBM model (94.9% accuracy) were superior to the random forest model with 89.1% accuracy. It is worth noting that the LightGBM had the highest accuracy of all models. This shows various subtle changes in models, illustrating the need for more analyses while assessing vehicle accidents. Machine learning is also a transforming tool in traffic safety analysis while providing vital guidelines for developing accurate traffic safety regulations.

show abstract

Section: Resultsmentioning

confidence: 99%

“…In addition, its precision, recall, and F1 metrics were highly competitive, demonstrating that it completely understood the differences between deadly and non-fatal patterns. PLOS ONE methods such as weighted quantile sketching [31,32]. This helps to strengthen the generalizability of the neural network.…”

Section: Xgboost Modelmentioning

confidence: 99%

Comparing fatal crash risk factors by age and crash type by using machine learning techniques

Alshehri,

Alanazi,

Yosri

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Their findings indicated that RFs perform best in fitted regressions, which is not consistent with our results. The reason for this phenomenon may be that we have a large amount of fixed sample data, and the XGBoost algorithm offers quicker training and prediction speed than RFs, which excel at obtaining predictions for smaller datasets (Vlachogiannis et al, 2023). Furthermore, machine learning also has many applications in carbon emissions prediction modeling, Zhao et al (2023) describes five different machine learning principles and their applications in various domains.…”

Section: Evaluation and Validation Of Machine Learning For Carbon Sto...mentioning

confidence: 99%

Estimating carbon sequestration potential and optimizing management strategies for Moso bamboo (Phyllostachys pubescens) forests using machine learning

Lv,

Yuan,

Sun

et al. 2024

Front. For. Glob. Change

View full text Add to dashboard Cite

Estimating the carbon sequestration potential of Moso bamboo (Phyllostachys pubescens) forests and optimizing management strategies play pivotal roles in enhancing quality and promoting sustainable development. However, there is a lack of methods to simulate changes in carbon sequestration capacity in Moso bamboo forests and to screen and optimize the best management measures based on long-term time series data from fixed-sample fine surveys. Therefore, this study utilized continuous survey data and climate data from fixed sample plots in Zhejiang Province spanning from 2004 to 2019. By comparing four different algorithms, namely random forest, support vector machine, XGBoost, and BP neural network, to construct aboveground carbon stock models for Moso bamboo forests. The ultimate goal was to identify the optimal algorithmic model. Additionally, the key driving parameters for future carbon stocks were considered and future aboveground carbon stocks were predicted in Moso bamboo forests. Then formulated an optimal management strategy based on these predictions. The results indicated that the carbon stock model constructed using the XGBoost algorithm, with an R2 of 0.9895 and root mean square error of 0.1059, achieved the best performance and was considered the optimal algorithmic model. The most influential driving parameters for vegetation carbon stocks in Moso bamboo forests were found to be mean age, mean diameter at breast height, and mean culm density. Under optimal management measures, which involve no harvesting of 1–3 du bamboo, 30% harvesting of 4 du bamboo, and 80% harvesting of bamboo aged 5 du and above. Our predictions show that aboveground carbon stocks in Moso bamboo forests in Zhejiang Province will peak at 36.25 ± 8.47 Tg C in 2046 and remain stable from 2046 to 2060. Conversely, degradation is detrimental to the long-term maintenance of carbon sequestration capacity in Moso bamboo forests, resulting in a peak aboveground carbon stock of 29.50 ± 7.49 Tg C in 2033, followed by a continuous decline. This study underscores the significant influence of estimating carbon sequestration potential and optimizing management decisions on enhancing and sustaining the carbon sequestration capacity of Moso bamboo forests.

show abstract

“…Emerging near real-time CV trajectory-level data provides far greater detail on individual passenger vehicle journey waypoints, and thus alleviates any over or underrepresentation concerns. Researchers have already demonstrated local or state-level examples of the versatility and sufficient representativeness of this CV trajectory data for use in assessing data coverage and filling gaps in traffic counts [29][30][31][32], monitoring mobility and safety through construction work zones [33][34][35][36][37], movement-level detection and performance monitoring at signalized intersections [38][39][40][41], and observing human mobility dynamics [42][43][44]. A pair of recent reports have built upon these methodologies and evaluated the usability of nationally available CV data sets at representative penetration rates towards analyzing the safety and mobility impacts of summer work zone construction as well as winter storm events on interstate travel in the United States [45].…”

Section: Emerging Connected Vehicle Data Opportunitiesmentioning

confidence: 99%

Analysis of Connected Vehicle Data to Quantify National Mobility Impacts of Winter Storms for Decision Makers and Media Reports

Desai,

Mathew,

et al. 2023

Future Transportation

View full text Add to dashboard Cite

Traditional techniques of monitoring roadway mobility during winter weather have relied on embedded road sensors, roadside cameras, radio reports from public safety staff, or public incident reports. However, widely available connected vehicle (CV) data provides government agencies and media with a unique opportunity to monitor the mobility impact of inclement weather events in near real-time. This study presents such a use case that analyzed over 500 billion CV records characterizing the spatial and temporal impact of a winter storm that moved across the country from 21 to 26 December 2022. The analysis covered 97,000 directional miles of interstate roadway and processed over 503 billion CV records. At the storm’s peak on 22 December at 5:26 PM Eastern Time, nearly 4800 directional miles of interstate roadway were operating under 45 mph, a widely accepted indicator of degraded interstate conditions. The study presents a methodological approach to systematically assess the mobility impact of this winter event on interstate roadways at a national and regional level. The paper then looks at a case study on Interstate 70, a 4350 directional mile route passing through ten states. Statewide comparison showed Ohio was most impacted, with 9% of mile-hours operating below 45 mph on 23 December. High-Resolution Rapid Refresh weather data provided by the National Oceanic and Atmospheric Administration was integrated into the analysis to provide a visualization of the storm’s temporal path and severity. We believe the proposed metrics and visualizations are effective tools for communicating the severity and geographic impact of extreme weather events to broad non-technical audiences.

show abstract

Intersense: An XGBoost model for traffic regulator identification at intersections through crowdsourced GPS data

Cited by 13 publications

References 28 publications

Comparing fatal crash risk factors by age and crash type by using machine learning techniques

Comparing fatal crash risk factors by age and crash type by using machine learning techniques

Estimating carbon sequestration potential and optimizing management strategies for Moso bamboo (Phyllostachys pubescens) forests using machine learning

Analysis of Connected Vehicle Data to Quantify National Mobility Impacts of Winter Storms for Decision Makers and Media Reports

Contact Info

Product

Resources

About