Search

Your search keyword '"Wu, Chaozhong"' showing total 464 results
Start Over Author "Wu, Chaozhong"
464 results on '"Wu, Chaozhong"'

3. A spatial-temporal short-term traffic flow prediction model based on dynamical-learning graph convolution mechanism

Author

Chen, Zhijun, Lu, Zhe, Chen, Qiushi, Zhong, Hongliang, Zhang, Yishi, Xue, Jie, and Wu, Chaozhong

Subjects
Computer Science - Machine Learning
Abstract

Short-term traffic flow prediction is a vital branch of the Intelligent Traffic System (ITS) and plays an important role in traffic management. Graph convolution network (GCN) is widely used in traffic prediction models to better deal with the graphical structure data of road networks. However, the influence weights among different road sections are usually distinct in real life, and hard to be manually analyzed. Traditional GCN mechanism, relying on manually-set adjacency matrix, is unable to dynamically learn such spatial pattern during the training. To deal with this drawback, this paper proposes a novel location graph convolutional network (Location-GCN). Location-GCN solves this problem by adding a new learnable matrix into the GCN mechanism, using the absolute value of this matrix to represent the distinct influence levels among different nodes. Then, long short-term memory (LSTM) is employed in the proposed traffic prediction model. Moreover, Trigonometric function encoding is used in this study to enable the short-term input sequence to convey the long-term periodical information. Ultimately, the proposed model is compared with the baseline models and evaluated on two real word traffic flow datasets. The results show our model is more accurate and robust on both datasets than other representative traffic prediction models., Comment: 21 pages, 16 figures

Published
2022

28. Feature Selection with Redundancy-complementariness Dispersion

Author

Chen, Zhijun, Wu, Chaozhong, Zhang, Yishi, Huang, Zhen, Ran, Bin, Zhong, Ming, and Lyu, Nengchao

Subjects
Computer Science - Learning, Statistics - Machine Learning, 68T10, 94A17, 62B10, 68U35, I.5.2, H.1.1
Abstract

Feature selection has attracted significant attention in data mining and machine learning in the past decades. Many existing feature selection methods eliminate redundancy by measuring pairwise inter-correlation of features, whereas the complementariness of features and higher inter-correlation among more than two features are ignored. In this study, a modification item concerning the complementariness of features is introduced in the evaluation criterion of features. Additionally, in order to identify the interference effect of already-selected False Positives (FPs), the redundancy-complementariness dispersion is also taken into account to adjust the measurement of pairwise inter-correlation of features. To illustrate the effectiveness of proposed method, classification experiments are applied with four frequently used classifiers on ten datasets. Classification results verify the superiority of proposed method compared with five representative feature selection methods., Comment: 28 pages, 13 figures, 7 tables

Published
2015

50. Analysis of Freeway Safety Influencing Factors on Driving Workload and Performance Based on the Gray Correlation Method

Author

Xie, Lian, Wu, Chaozhong, Duan, Min, and Lyu, Nengchao

Subjects
Traffic safety -- Methods, Traffic accidents -- Methods, Transportation industry
Abstract

Human-related factors are a crucial inducement of traffic accidents. Understanding the influence of freeway environments on the driving behavior and workload experienced by drivers has been demonstrated to be of primary importance for improving traffic safety. To study the effect of alignment, traffic flow, and sign information on drivers' mental workload and behavior, 16 scenarios were constructed using the orthogonal design method, and simulated driving experiments were carried out with 45 participants. During driving, indicators such as the mean and standard deviation of vehicle speed and lane departure were collected, and the NASA-TLX questionnaire was adopted to measure workload. Analysis of variance results indicated that the radius of the horizontal curve, gradient, flow, and sign information level have a significant influence on drivers' workload and speed keeping ability. In addition, the horizontal curve radius has a significant effect on lane keeping ability. The importance of safety influencing factors on driving workload and performance was quantitatively ranked by integrating the trend of Deng's correlation degree, comprehensive correlation degree, and similar correlation degree, whose weight was calculated using the entropy method. Traffic sign information was found to have the greatest impact on workload. In terms of driving performance, traffic volume has the greatest influence on the mean and standard deviation of vehicle speed, followed by the amount of sign information. Lane departure is most affected by the radius of the horizontal curve. These findings provide guidance for freeway traffic safety regulation, including workload control and road facility optimization., Author(s): Lian Xie [1,2]; Chaozhong Wu [1,3]; Min Duan [4]; Nengchao Lyu (corresponding author) [1,3] 1. Introduction Statistics of road traffic accidents show that 80% of accidents and 65% of [...]

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results