Combining probabilistic neural networks and decision trees for maximally accurate and efficient accident prediction

Tambouratzis, Tatiana; Souliou, Dora; Chalikias, Miltiadis; Gregoriades, Andreas

doi:10.1109/ijcnn.2010.5596610

Cited by 14 publications

(6 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two-layered control system lets implement random movements of robot in supervisor mode. The structure of the control system assumes its development with an increase of amount and quality of information about the operating environment [11].…”

Section: Resultsmentioning

confidence: 99%

The Control System of the Eight-Legged Mobile Walking Robot

Andreev

Жога

Serov

et al. 2014

Communications in Computer and Information Science

View full text Add to dashboard Cite

Abstract. The paper considers the control system of the autonomous mobile robot CS-7 with eight supporting legs, installed on four hard frames in pairs. Advantages of the chosen design: energy absence to support machine weight and simpler algorithms of trajectory and reference points motion management.Onboard control is designed as a scalable, modular distributed low-level microprocessor system, which comprises two major subsystems: information measurement for processing sensors data and robot actuators control.Supervisory control mode is supported by the vision system based on timeof-flight cameras and a set of ultrasonic and infrared sensors, geosteering module for local and global robot positioning. Measuring information is transmitted to the control computer via Wi-Fi module. Software for the upper-level computer and library of functions for low-level units are developed for the implementation of various robot motion algorithms. Functions define the basic parameters of orthogonal movers depending on the given route and environment map.

show abstract

Section: Resultsmentioning

confidence: 99%

The Control System of the Eight-Legged Mobile Walking Robot

Andreev

Жога

Serov

et al. 2014

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…Tambouratzis et al [22] used a combination of artificial neural networks and decision trees to predict the severity (mild, severe, or fatal) of accidents. The data used in that study refer to accidents in Cyprus during 2005.…”

Section: Classification Of Accidents Using Machine Learning Techniquesmentioning

confidence: 99%

“…Proposed Method Neural Network 83.00% 82.00% Tiwari et al [37] Lazy Classifier (IBK) 84.47% -Kumar et al [8] Random Forest 81% -Satu et al [23] J48 (pruned) 78.9% -Iranitalab and Khattak [24] KNN and K-means 73.95% -Tambouratzis et al [22] Neural Network and DecisionTree 70% -Bülbül et al [19] CART -81.1% Tiwari et al [38] Decision Tree -71.8% Zhang et al [39] K-means and Bayesian network -59% Yu et al [40] Neural Network -73.65%…”

Section: Algorithms Acc F1mentioning

confidence: 99%

A Machine Learning Approach for Classifying Road Accident Hotspots

Amorim

Firmino

Baptista

et al. 2023

IJGI

View full text Add to dashboard Cite

Road accidents are a worldwide problem, affecting millions of people annually. One way to reduce such accidents is to predict risk areas and alert drivers. Advanced research has been carried out on identifying accident-influencing factors and potential highway risk areas to mitigate the number of road accidents. Machine learning techniques have been used to build prediction models using a supervised classification based on a labeled dataset. In this work, we experimented with many machine learning algorithms to discover the best classifier for the Brazilian federal road hotspots associated with severe or nonsevere accident risk using several features. We tested with SVM, random forest, and a multi-layer perceptron neural network. The dataset contains a ten-year road accident report by the Brazilian Federal Highway Police. The feature set includes spatial footprint, weekday and time when the accident happened, road type, route, orientation, weather conditions, and accident type. The results were promising, and the neural network model provided the best results, achieving an accuracy of 83%, a precision of 84%, a recall of 83%, and an F1-score of 82%.

show abstract

“…Initially, the key factors involved in road collisions were identified, similarly to [10], and are summarised below.…”

Section: Accident Data Setmentioning

confidence: 99%

“…In order to measure the error of the models, we utilised Chebyshev distance, calculated by Equation (10). Chebyshev distance between two vectors, in this case the predicted and actual number of accidents, is the greatest distance in any state.…”

Section: Mapping Bayesian Network To Neural Networkmentioning

confidence: 99%

On making sense of neural networks in road analysis

Morris

Αντωνιάδης

Took

2017

2017 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Abstract-Neural networks have been treated as "black boxes" for the majority of the machine learning community. The difficulty in making sense of neural networks lies in the complex topology of the hidden layers. Although there have been works in the literature aimed at demystifying the way neural networks operate, making sense of the hidden layer still remains a challenge. In this work, we propose a way to derive physical meaning from the hidden layer by mapping our neural network to the topology of a Bayesian network. Using this mapping, we enhance the probabilities of the Bayesian network resulting in a hybrid model that outperforms both the Bayesian and neural networks in the task of traffic accident prediction. Our analysis suggests that a neural network can estimate the node probabilities of a Bayesian network if mapped accordingly.

show abstract

Combining probabilistic neural networks and decision trees for maximally accurate and efficient accident prediction

Cited by 14 publications

References 24 publications

The Control System of the Eight-Legged Mobile Walking Robot

The Control System of the Eight-Legged Mobile Walking Robot

A Machine Learning Approach for Classifying Road Accident Hotspots

On making sense of neural networks in road analysis

Contact Info

Product

Resources

About