Development of Driver-Behavior Model Based onWOA-RBM Deep Learning Network

Liu, Junhui; Jia, Yang; Wang, Yaya; Doležel, Petr

doi:10.1155/2020/8859891

Cited by 8 publications

(8 citation statements)

References 35 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To recognize the joint distribution of obvious and hidden layers an energy function was used which follows the following formula 39 :

P false(v, h false) = \frac{e^{- E (v, h)}}{\sum_{v, h} e^{- E false(v, h false)}},

where,

E false(v, h false)

defines the energy function of RBM and can be gained as follows 40 :

E false(v, h false) = - \sum_{i = 1} a_{i} v_{i} - \sum_{j = 1} b_{j} h_{j} - \sum_{i, j} v_{i} h_{j} w_{italicij},

where,

w_{italicij}

is the weight between the layer of obvious and the layer of hidden, and

a_{i}

and

b_{j}

represent the coefficients for the obvious and the hidden layers, respectively.…”

Section: Methodsmentioning

confidence: 99%

“…This is accomplished with a desirable choice of the factors a, b, and gain for the RBM. By analyzing a given sample can rewrite the probability as follows 40

P false(v false) = \frac{\sum_{v} e^{- E false(v, h false)}}{\sum_{v, h} e^{- E false(v, h false)}} .

Through utilizing the random gradient rise for the derivation of the logarithm,

w_{t + 1} = w_{t} + ϑ (P (false(h false| v false) v^{T}) - P (false(\overset{h}{true} false| \overset{v}{true} false) {\overset{v}{true}}^{T})) - {λ w}_{t} + α Δ w_{t - 1},

a_{t + 1} = a_{t} + ϑ false(v - \overset{true^}{v} false) + α Δ a_{t - 1},

b_{t + 1} = b_{t} + ϑ false(P false(h false| v false) - P false(\overset{h}{true} false| \overset{v}{true} false) false) + α Δ b_{t - 1},

where,

P (h_{j} = ...)

…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Using an optimized soil and water assessment tool by deep belief networks to evaluate the impact of land use and climate change on water resources

Khayatnezhad

Youssefi

2022

Concurrency and Computation

View full text Add to dashboard Cite

This article investigates the negative effect of land use and climate changes on water resources by the SWAT and SWAT-DEEP\LMSFO model. Due to the importance of runoff impact on water resources in this article, the hybrid hydrological-deep neural networks optimized by the improved SFO based on logistic map (LMSFO) algorithm have been used to provide more accurate results for runoff estimation. This method improves runoff simulation. Firstly, runoff under the influence of land use and climate change is estimated by the SWAT model. Once again, runoff is estimated by the SWAT-DEEP/LMSFO model. In the DEEP/LMSFO model, the primary runoff has been estimated by an un-calibrated SWAT model. Then the Primary runoff simulated is entered as an input into the DEEP model. Finally, the runoff is estimated by a test-training method. The results of the SWAT model and the DEEP/LMSFO model show that there is an inverse correlation between land use so that reducing land cover increases runoff. The results show that climate change and land-use change can affect annual runoff changes in the coming years.

show abstract

“…To recognize the joint distribution of obvious and hidden layers an energy function was used which follows the following formula 39 :

P false(v, h false) = \frac{e^{- E (v, h)}}{\sum_{v, h} e^{- E false(v, h false)}},

where,

E false(v, h false)

defines the energy function of RBM and can be gained as follows 40 :

E false(v, h false) = - \sum_{i = 1} a_{i} v_{i} - \sum_{j = 1} b_{j} h_{j} - \sum_{i, j} v_{i} h_{j} w_{italicij},

where,

w_{italicij}

is the weight between the layer of obvious and the layer of hidden, and

a_{i}

and

b_{j}

represent the coefficients for the obvious and the hidden layers, respectively.…”

Section: Methodsmentioning

confidence: 99%

“…This is accomplished with a desirable choice of the factors a, b, and gain for the RBM. By analyzing a given sample can rewrite the probability as follows 40

P false(v false) = \frac{\sum_{v} e^{- E false(v, h false)}}{\sum_{v, h} e^{- E false(v, h false)}} .

Through utilizing the random gradient rise for the derivation of the logarithm,

w_{t + 1} = w_{t} + ϑ (P (false(h false| v false) v^{T}) - P (false(\overset{h}{true} false| \overset{v}{true} false) {\overset{v}{true}}^{T})) - {λ w}_{t} + α Δ w_{t - 1},

a_{t + 1} = a_{t} + ϑ false(v - \overset{true^}{v} false) + α Δ a_{t - 1},

b_{t + 1} = b_{t} + ϑ false(P false(h false| v false) - P false(\overset{h}{true} false| \overset{v}{true} false) false) + α Δ b_{t - 1},

where,

P (h_{j} = ...)

…”

Section: Methodsmentioning

confidence: 99%

Using an optimized soil and water assessment tool by deep belief networks to evaluate the impact of land use and climate change on water resources

Khayatnezhad

Youssefi

2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…The authors of the paper [24] used the clustering technique, machine learning algorithms and deep learning algorithms to classify drivers' behaviors into those eco-friendly and those which are not. In the paper [25], deep learning was used as one of the subfields of machine learning for modelling drivers' behavior. The most frequently used types of machine learning algorithms in the papers are SVM (Support Vector Machines), NN (Neural Networks), BL (Bayesian Learners) and EL (Ensemble Learners), whereas the algorithms of the Decision Trees (DT) types and Instance Based (IB) algorithms are present to a much lesser extent [26].…”

Section: Machine Learning Methodsmentioning

confidence: 99%

Evaluation of the influence of terrain and traffic road conditions on the driver’s driving performances by applying machine learning

et al. 2022

View full text Add to dashboard Cite

In this paper, research is done in the influence of different terrain and traffic conditions on road sections on the driver?s driving performances, i.e. on the car energy efficiency and CO2 emission. A methodology aimed at determining to which extent unfavorable traffic and/or terrain conditions on a road section contribute to the driver?s worse driving performances, and also to determine when the driver?s aggressive driving style is responsible for greater fuel consumption and greater CO2 emission is proposed. In order to apply the proposed methodology, a research study was carried out in a cargo transportation company and 12 drives who drove the same vehicle on five different road sections were selected. As many as 284 014 of the instances of the data about the defined parameters of the road section and the driver?s driving style were collected, based on which and with the help of machine learning a prediction of the scores for the road section and the scores for the driver?s driving style was performed. The obtained results have shown that the proposed methodology is a useful tool for managers enabling them to simply and quickly determine potential room for increasing the energy efficiency of the vehicle fleet and decreasing CO2 emission.

show abstract

“…These make WOA an ideal candidate for resolving constrained and unconstrained optimization problems delivering real-time applications without much of a structural reformation of the existing algorithm. The study in [22] developed a WOA-based Restricted Boltzmann machine (RBM) neural network framework considering real-time driver behavior data. The simulation results generated an accuracy of 90% representing human drivers' operation effectively and efficiently.…”

Section: Whale Optimization Algorithm (Woa)mentioning

confidence: 99%

Whale Optimization Algorithm-Based Deep Learning Model for Driver Identification in Intelligent Transport Systems

Li¹,

Sun²,

Hu³

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Driver identification in intelligent transport systems has immense demand, considering the safety and convenience of traveling in a vehicle. The rapid growth of driver assistance systems (DAS) and driver identification system propels the need for understanding the root causes of automobile accidents. Also, in the case of insurance, it is necessary to track the number of drivers who commonly drive a car in terms of insurance pricing. It is observed that drivers with frequent records of paying "fines" are compelled to pay higher insurance payments than drivers without any penalty records. Thus driver identification act as an important information source for the intelligent transport system. This study focuses on a similar objective to implement a machine learning-based approach for driver identification. Raw data is collected from in-vehicle sensors using the controller area network (CAN) and then converted to binary form using a one-hot encoding technique. Then, the transformed data is dimensionally reduced using the Principal Component Analysis (PCA) technique, and further optimal parameters from the dataset are selected using Whale Optimization Algorithm (WOA). The most relevant features are selected and then fed into a Convolutional Neural Network (CNN) model. The proposed model is evaluated against four different use cases of driver behavior. The results show that the best prediction accuracy is achieved in the case of drivers without glasses. The proposed model yielded optimal accuracy when evaluated against the K-Nearest Neighbors (KNN) and Support Vector Machines (SVM) models with and without using dimensionality reduction approaches.

show abstract

Development of Driver-Behavior Model Based onWOA-RBM Deep Learning Network

Cited by 8 publications

References 35 publications

Using an optimized soil and water assessment tool by deep belief networks to evaluate the impact of land use and climate change on water resources

Using an optimized soil and water assessment tool by deep belief networks to evaluate the impact of land use and climate change on water resources

Evaluation of the influence of terrain and traffic road conditions on the driver’s driving performances by applying machine learning

Whale Optimization Algorithm-Based Deep Learning Model for Driver Identification in Intelligent Transport Systems

Contact Info

Product

Resources

About