Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.237872
Ya-Xin Zhou, Shiyuan Han, Jin Zhou, Kang Yao
Quality evaluation for higher engineering education has important guiding significance and feedback role on cultivating engineering talents. Combining with the educational core concept of outcomes-based education (OBE) and the educational process data, a fuzzy quality evaluation algorithm is developed for engineering education deriving from a constructed Quasi-Neural-Network (QNN) framework. More specifically, considering the logical relationships among basic components in the whole process of engineering education, a four-layers QNN framework is designed first to underly and implement the educational concept of OBE reasonably, which includes the training objectives layer, requirement capability for graduation layer, requirement sub-capability for graduation layer, and course layer. After that, by employing the educational process data under the proposed QNN framework, a fuzzy comprehensive evaluation algorithm is designed to describe the achievement scale of target capability for engineering education. Finally, focusing on the research capability for computer science with related four courses, the experiments based on the process educational data sets show the superiority and efficiency of the proposed framework and algorithm.
{"title":"Fuzzy Quality Evaluation Algorithm for Higher Engineering Education Quality via Quasi-neural-network Framework","authors":"Ya-Xin Zhou, Shiyuan Han, Jin Zhou, Kang Yao","doi":"10.1109/SPAC49953.2019.237872","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237872","url":null,"abstract":"Quality evaluation for higher engineering education has important guiding significance and feedback role on cultivating engineering talents. Combining with the educational core concept of outcomes-based education (OBE) and the educational process data, a fuzzy quality evaluation algorithm is developed for engineering education deriving from a constructed Quasi-Neural-Network (QNN) framework. More specifically, considering the logical relationships among basic components in the whole process of engineering education, a four-layers QNN framework is designed first to underly and implement the educational concept of OBE reasonably, which includes the training objectives layer, requirement capability for graduation layer, requirement sub-capability for graduation layer, and course layer. After that, by employing the educational process data under the proposed QNN framework, a fuzzy comprehensive evaluation algorithm is designed to describe the achievement scale of target capability for engineering education. Finally, focusing on the research capability for computer science with related four courses, the experiments based on the process educational data sets show the superiority and efficiency of the proposed framework and algorithm.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114463628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.237865
Nan Miao, Shengjie Zhao, Qingjiang Shi, Rongqing Zhang
With the emerging of smart metering around the world, there is a growing demand to analyse the residential energy usage. In this paper, we propose a Deep Neural Network (DNN)-based approach for non-intrusive load monitoring (NILM), which can achieve effective and efficient estimation of individual appliance usage according to a single main meter reading in a non-intrusive manner. Considering practical situations, two training methods are provided. The first training approach is fully supervised learning, which requires a ground truth of label, indicating the state of the appliance (ON/OFF), to build a prediction model. The second training approach is semi-supervised learning, leading to better performance by F-Measure metric while only requiring some more unlabelled training data. Experimental results on the low-sample rate REDD dataset demonstrate the superior performance of our proposed DNN-based method compared with Hidden Markov Model (HMM)based and Graph Signal Processing (GSP)-based approaches.
{"title":"Non-Intrusive Load Disaggregation Using Semi-Supervised Learning Method","authors":"Nan Miao, Shengjie Zhao, Qingjiang Shi, Rongqing Zhang","doi":"10.1109/SPAC49953.2019.237865","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237865","url":null,"abstract":"With the emerging of smart metering around the world, there is a growing demand to analyse the residential energy usage. In this paper, we propose a Deep Neural Network (DNN)-based approach for non-intrusive load monitoring (NILM), which can achieve effective and efficient estimation of individual appliance usage according to a single main meter reading in a non-intrusive manner. Considering practical situations, two training methods are provided. The first training approach is fully supervised learning, which requires a ground truth of label, indicating the state of the appliance (ON/OFF), to build a prediction model. The second training approach is semi-supervised learning, leading to better performance by F-Measure metric while only requiring some more unlabelled training data. Experimental results on the low-sample rate REDD dataset demonstrate the superior performance of our proposed DNN-based method compared with Hidden Markov Model (HMM)based and Graph Signal Processing (GSP)-based approaches.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125852570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.237882
Qiushi Tian, Jin Zhou, Shiyuan Han, Lin Wang, Yuehui Chen
Traditional kernel clustering methods are useful in dealing with non-linear data, but the high-dimensional kernel space obtained by kernel mapping is an abstract concept, which is difficult to be determined. The kernel mapping between raw data space and kernel space needs high computational complexity which is burdensome for hardware. At the same time, due to the unknown nature of kernel space, traditional kernel clustering methods cannot process data with the consideration of different importance among dimensions, i.e., discover the hidden feature subset of high-dimensional sparse data. To overcome these limitations, we put forward a novel random Fourier feature based attribute-weighed kernel fuzzy c-means clustering algorithm (RFF-WKFCM). This method employs RFF map to generate low-rank random features, and performs fuzzy c-means clustering with attribute weight entropy regularization in this feature space, which greatly reduces the computational complexity. What is more, the adoption of the maximum entropy technique ensures the optimal distribution of attribute weights, which stimulate important dimensions play a greater role in the clustering process. The proposed method shows good performance on the experiments of ring data set compared with other fuzzy clutering methods.
{"title":"Random Feature Based Attribute-weighed Kernel Fuzzy Clustering for Non-linear Data","authors":"Qiushi Tian, Jin Zhou, Shiyuan Han, Lin Wang, Yuehui Chen","doi":"10.1109/SPAC49953.2019.237882","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237882","url":null,"abstract":"Traditional kernel clustering methods are useful in dealing with non-linear data, but the high-dimensional kernel space obtained by kernel mapping is an abstract concept, which is difficult to be determined. The kernel mapping between raw data space and kernel space needs high computational complexity which is burdensome for hardware. At the same time, due to the unknown nature of kernel space, traditional kernel clustering methods cannot process data with the consideration of different importance among dimensions, i.e., discover the hidden feature subset of high-dimensional sparse data. To overcome these limitations, we put forward a novel random Fourier feature based attribute-weighed kernel fuzzy c-means clustering algorithm (RFF-WKFCM). This method employs RFF map to generate low-rank random features, and performs fuzzy c-means clustering with attribute weight entropy regularization in this feature space, which greatly reduces the computational complexity. What is more, the adoption of the maximum entropy technique ensures the optimal distribution of attribute weights, which stimulate important dimensions play a greater role in the clustering process. The proposed method shows good performance on the experiments of ring data set compared with other fuzzy clutering methods.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114863847","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.237869
Zhiyu Chen, Yangwei Gu, Chunhua Deng, Ziqi Zhu
Learning deep representations have been widely used in action recognition task. However, the features extracted by deep convolutional neural networks (CNNs) have many redundant information. This paper aims to discover the relevance between temporal features extracted by CNNs. Different fromTemporal Segment Networks (TSN) to randomly select video clips. Based on the matrix-based Rényi’s α-entropy, we estimate the mutual information between temporal domain features. Through our experiments, we propose an adaptive temporal segmentation scheme to represent the entire videos. We also combine the features of RGB and optical flow frames extracted by 3D ConvNets to verify the complementary information between them. We show that the proposed approach achieves 94.4 and 72.8 percent accuracy, in the UCF- 101 and HMDB-51 datasets.
{"title":"Adaptive Temporal Segmentation for Action Recognition","authors":"Zhiyu Chen, Yangwei Gu, Chunhua Deng, Ziqi Zhu","doi":"10.1109/SPAC49953.2019.237869","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237869","url":null,"abstract":"Learning deep representations have been widely used in action recognition task. However, the features extracted by deep convolutional neural networks (CNNs) have many redundant information. This paper aims to discover the relevance between temporal features extracted by CNNs. Different fromTemporal Segment Networks (TSN) to randomly select video clips. Based on the matrix-based Rényi’s α-entropy, we estimate the mutual information between temporal domain features. Through our experiments, we propose an adaptive temporal segmentation scheme to represent the entire videos. We also combine the features of RGB and optical flow frames extracted by 3D ConvNets to verify the complementary information between them. We show that the proposed approach achieves 94.4 and 72.8 percent accuracy, in the UCF- 101 and HMDB-51 datasets.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129042507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.243776
Xuchen Wang, Yuxuan Huang, Dengxiu Yu, Mingyong Liu
The paper proposes the self-organized clustering-fission swarm system based on the coupling degree of weighted information. In previous work, researchers study the clustering or fission based on information coupling degree. However, the performance of clustering-fission is effected by the designing formation coupling degree. Adding the weight into information coupling degree can improve the performance of clustering-fission. The bigger the weight is, the higher the probability of clustering-fission occurrence will become. One main contribution of this paper is adjusted by weight. Finally, the proposed method is verified by simulation.
{"title":"Self-organized Clustering -Fission Swarm System Based on the Coupling Degree of Weighted Information","authors":"Xuchen Wang, Yuxuan Huang, Dengxiu Yu, Mingyong Liu","doi":"10.1109/SPAC49953.2019.243776","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.243776","url":null,"abstract":"The paper proposes the self-organized clustering-fission swarm system based on the coupling degree of weighted information. In previous work, researchers study the clustering or fission based on information coupling degree. However, the performance of clustering-fission is effected by the designing formation coupling degree. Adding the weight into information coupling degree can improve the performance of clustering-fission. The bigger the weight is, the higher the probability of clustering-fission occurrence will become. One main contribution of this paper is adjusted by weight. Finally, the proposed method is verified by simulation.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124168415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.237883
Lei Li, Wei Li, Jianxing Liao, Xuegang Hu
Compared with univariate time series clustering, multivariate time series (MTS) clustering has become a challenging research topic on the data mining of time series. In this paper, a novel model-based approach Adaptive State Continuity-Based Sparse Inverse Covariance Clustering (ASCSICC) is proposed for MTS clustering. Specifically, the state continuity is introduced to make the traditional Gaussian mixture model (GMM) applicable to time series clustering. To prevent overfitting, the alternating direction method of multipliers (ADMM) is applied to optimize the parameter of GMM inverse covariance. In addition, the proposed approach simultaneously segments and clusters the time series. Technically, first, the adaptive state continuity is estimated based on the distance similarity of adjacent time series data. Then, a dynamic programming algorithm of cluster assignment by adaptive state continuity is taken as the E-step, and the ADMM for solving sparse inverse covariance is taken as the M-step. E-step and M-step are combined into an Expectation-Maximization (EM) algorithm to conduct the clustering process. Finally, we show the effectiveness of the proposed approach by comparing the ASC-SICC with several state-of-the-art approaches in experiments on two datasets from real applications.
{"title":"Adaptive State Continuity-Based Sparse Inverse Covariance Clustering for Multivariate Time Series","authors":"Lei Li, Wei Li, Jianxing Liao, Xuegang Hu","doi":"10.1109/SPAC49953.2019.237883","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237883","url":null,"abstract":"Compared with univariate time series clustering, multivariate time series (MTS) clustering has become a challenging research topic on the data mining of time series. In this paper, a novel model-based approach Adaptive State Continuity-Based Sparse Inverse Covariance Clustering (ASCSICC) is proposed for MTS clustering. Specifically, the state continuity is introduced to make the traditional Gaussian mixture model (GMM) applicable to time series clustering. To prevent overfitting, the alternating direction method of multipliers (ADMM) is applied to optimize the parameter of GMM inverse covariance. In addition, the proposed approach simultaneously segments and clusters the time series. Technically, first, the adaptive state continuity is estimated based on the distance similarity of adjacent time series data. Then, a dynamic programming algorithm of cluster assignment by adaptive state continuity is taken as the E-step, and the ADMM for solving sparse inverse covariance is taken as the M-step. E-step and M-step are combined into an Expectation-Maximization (EM) algorithm to conduct the clustering process. Finally, we show the effectiveness of the proposed approach by comparing the ASC-SICC with several state-of-the-art approaches in experiments on two datasets from real applications.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127367252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.244103
J. Fan, Lin Wang, Chen Xiao, Bo Yang, Jin Zhou
Texture is of great significance to the study of cement field. It can reflect various information, such as cement strength and hydration age. However, the texture of cement hydration image is complex and diverse, and most of the methods are relatively inefficient at present. Therefore, we propose a fast way to synthesize texture through neural network. It uses the information of the causal neighborhood to extract their implicit features. This method is more perfect than the simple expression method, and can extract more implicit features and get a better neural network model. Through this model we can quickly and easily synthesize cement texture images. This algorithm is faster than the current popular methods and more diverse than the methods of gene expression programming.
{"title":"Cement Texture Synthesis Based on Feedforward Neural Network","authors":"J. Fan, Lin Wang, Chen Xiao, Bo Yang, Jin Zhou","doi":"10.1109/SPAC49953.2019.244103","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.244103","url":null,"abstract":"Texture is of great significance to the study of cement field. It can reflect various information, such as cement strength and hydration age. However, the texture of cement hydration image is complex and diverse, and most of the methods are relatively inefficient at present. Therefore, we propose a fast way to synthesize texture through neural network. It uses the information of the causal neighborhood to extract their implicit features. This method is more perfect than the simple expression method, and can extract more implicit features and get a better neural network model. Through this model we can quickly and easily synthesize cement texture images. This algorithm is faster than the current popular methods and more diverse than the methods of gene expression programming.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115591446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thanks to the large-scale smart meters deployments around the world, non-intrusive appliance load monitoring (NILM) is receiving popularity. It aims to disaggregate the total electricity load of a home into individual appliances without resorting to any specific appliance power monitors. NILM is worthy of broad attention owing to its facilitation in energy savings. This paper regards NILM as a classification task and proposes a two-step method based on graph signal processing (GSP). In the first step, a smoothest solution is obtained by minimizing the regularization term. In the second step, gradient projection method, which uses the obtained minimizer as a start point, is adopted to optimize the while objective function, where NILM is regarded as a constrained nonlinear programming problem. The experiment results based on the open-access data set REDD clearly demonstrate that the proposed GSP-based method achieves improved performance compared with other state-of-the-art low-rate NILM approaches.
{"title":"Low-Rate Non-Intrusive Appliance Load Monitoring Based on Graph Signal Processing","authors":"Bing Zhang, Shengjie Zhao, Qingjiang Shi, Rongqing Zhang","doi":"10.1109/SPAC49953.2019.237866","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237866","url":null,"abstract":"Thanks to the large-scale smart meters deployments around the world, non-intrusive appliance load monitoring (NILM) is receiving popularity. It aims to disaggregate the total electricity load of a home into individual appliances without resorting to any specific appliance power monitors. NILM is worthy of broad attention owing to its facilitation in energy savings. This paper regards NILM as a classification task and proposes a two-step method based on graph signal processing (GSP). In the first step, a smoothest solution is obtained by minimizing the regularization term. In the second step, gradient projection method, which uses the obtained minimizer as a start point, is adopted to optimize the while objective function, where NILM is regarded as a constrained nonlinear programming problem. The experiment results based on the open-access data set REDD clearly demonstrate that the proposed GSP-based method achieves improved performance compared with other state-of-the-art low-rate NILM approaches.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130306608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.243772
Bianna Chen, Tong Zhang, Xue Jia, Jianxiu Jin, C. L. P. Chen, Xiangmin Xu
Feature selection is used to extract the most essential features from the data without degrading the performance of an algorithm, especially a classification algorithm. Various evolutionary algorithms (EAs) combined with classification algorithms are commonly used for feature selection. This paper suggests an innovative feature selection algorithm based on I-Ching Divination Evolutionary Algorithm, called binary IDEA (BIDEA). The main idea is to use a series of hexagrams encoded as binary vectors, which is called the hexagram state to represent the solutions of selected features. After three flexible operations, intrication, turnover and mutual, the transformed hexagram state can be obtained as candidate solutions. Then the optimized hexagram state can be searched to form the new state in the next iteration by evaluating candidate solutions. Experiments checked out with standard datasets reveal that the proposed BIDEA performs better in terms of classification accuracy, precision, recall and feature reduction than the competing feature selection methods.
{"title":"A Binary I-Ching Divination Evolutionary Algorithm for Feature Selection","authors":"Bianna Chen, Tong Zhang, Xue Jia, Jianxiu Jin, C. L. P. Chen, Xiangmin Xu","doi":"10.1109/SPAC49953.2019.243772","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.243772","url":null,"abstract":"Feature selection is used to extract the most essential features from the data without degrading the performance of an algorithm, especially a classification algorithm. Various evolutionary algorithms (EAs) combined with classification algorithms are commonly used for feature selection. This paper suggests an innovative feature selection algorithm based on I-Ching Divination Evolutionary Algorithm, called binary IDEA (BIDEA). The main idea is to use a series of hexagrams encoded as binary vectors, which is called the hexagram state to represent the solutions of selected features. After three flexible operations, intrication, turnover and mutual, the transformed hexagram state can be obtained as candidate solutions. Then the optimized hexagram state can be searched to form the new state in the next iteration by evaluating candidate solutions. Experiments checked out with standard datasets reveal that the proposed BIDEA performs better in terms of classification accuracy, precision, recall and feature reduction than the competing feature selection methods.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123189626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-01DOI: 10.1109/SPAC49953.2019.237874
Xiao-Yue Ma, Ya Fang, Shiyuan Han, Ya-Xin Zhou, Ke Yang, Jin Zhou, Kang Yao
Artificial intelligence has been widely used in traffic flow prediction. In this paper, we investigate how the seemingly disorganized behavior of traffic flow prediction could be well represented by using flexible neural tree (FNT).The traffic flow data of the previous two months were analyzed and trained to construct a flexible neural tree model. This paper investigates the changing law of traffic volume and makes scientific and reasonable prediction for future traffic volume. By using particle swarm optimization (PSO) algorithm to optimize the parameters of FNT to build a better prediction model. The proposed method has good adaptability and robustness. It can provide a reliable model for traffic flow prediction. According to the experimental results, the prediction model can accurately describe the changing trend of traffic flow.
{"title":"Traffic Flow Prediction Algorithm Based on Flexible Neural Tree","authors":"Xiao-Yue Ma, Ya Fang, Shiyuan Han, Ya-Xin Zhou, Ke Yang, Jin Zhou, Kang Yao","doi":"10.1109/SPAC49953.2019.237874","DOIUrl":"https://doi.org/10.1109/SPAC49953.2019.237874","url":null,"abstract":"Artificial intelligence has been widely used in traffic flow prediction. In this paper, we investigate how the seemingly disorganized behavior of traffic flow prediction could be well represented by using flexible neural tree (FNT).The traffic flow data of the previous two months were analyzed and trained to construct a flexible neural tree model. This paper investigates the changing law of traffic volume and makes scientific and reasonable prediction for future traffic volume. By using particle swarm optimization (PSO) algorithm to optimize the parameters of FNT to build a better prediction model. The proposed method has good adaptability and robustness. It can provide a reliable model for traffic flow prediction. According to the experimental results, the prediction model can accurately describe the changing trend of traffic flow.","PeriodicalId":410003,"journal":{"name":"2019 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123661946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：