Pub Date : 2012-05-29DOI: 10.1109/ICNC.2012.6234620
Yuhua Liu, P. Xie, Hongxia Liu
On the basis of the establishment of evaluation index for book procurement tender, an evaluating system of book procurement tender was presented based on BP neural network algorithm. This designed system increased the speed of algorithm and improved the performance of algorithm. The analytic hierarchy process method was used to generate samples of network, which used the advantage of BP neural networks effectively and avoided some human errors in the process of evaluation for book procurement tender. Simulation results showed that this system was satisfactory. It can overcome some disturbances coming from subjective determine of normal values and weights effectively in the process of evaluation and evaluate the suppliers objectively. Therefore, the system can be an effective tool for choosing proper suppliers during the book procurement tender in the library.
{"title":"An evaluating system for invite bid of Chinese Books purchasing based on BP neural network","authors":"Yuhua Liu, P. Xie, Hongxia Liu","doi":"10.1109/ICNC.2012.6234620","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234620","url":null,"abstract":"On the basis of the establishment of evaluation index for book procurement tender, an evaluating system of book procurement tender was presented based on BP neural network algorithm. This designed system increased the speed of algorithm and improved the performance of algorithm. The analytic hierarchy process method was used to generate samples of network, which used the advantage of BP neural networks effectively and avoided some human errors in the process of evaluation for book procurement tender. Simulation results showed that this system was satisfactory. It can overcome some disturbances coming from subjective determine of normal values and weights effectively in the process of evaluation and evaluate the suppliers objectively. Therefore, the system can be an effective tool for choosing proper suppliers during the book procurement tender in the library.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131382902","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234505
Peng Zhang, Yu-tong Chang
Considering determining the number of software fault is an uncertain non-linear problem with only small sample, a novel software fault prediction method based on grey neural network is put forward. Firstly, constructing the grey neural network topological structure according the small sample sequence is necessary, and then the network learning algorithm is discussed. Finally, the grey neural network prediction model based on the grey theory and artificial neural network is proposed. The sample fault sequences of some software project are used to verify the precision of this method. Comparison with GM(1,1), the proposed model can reduce the prediction relative error effectively.
{"title":"Software fault prediction based on grey neural network","authors":"Peng Zhang, Yu-tong Chang","doi":"10.1109/ICNC.2012.6234505","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234505","url":null,"abstract":"Considering determining the number of software fault is an uncertain non-linear problem with only small sample, a novel software fault prediction method based on grey neural network is put forward. Firstly, constructing the grey neural network topological structure according the small sample sequence is necessary, and then the network learning algorithm is discussed. Finally, the grey neural network prediction model based on the grey theory and artificial neural network is proposed. The sample fault sequences of some software project are used to verify the precision of this method. Comparison with GM(1,1), the proposed model can reduce the prediction relative error effectively.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131102066","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234764
Shizhong Liao, Menghua Duan
Sketch recognition is one of the essential step of sketch understanding. Challenge in sketch recognition is the variation and imprecision present in sketch. Free drawing styles of sketching make it difficult to build a robust sketch recognition system. This paper proposes a novel recognition approach that can recognize primitive shapes, as well as combinations of these primitives. The approach is independent of stroke order, number, as well as invariant to size and aspect ratio of sketch. Feature string is used to represent primitives. We defined a similarity measure on these feature strings that counts common substrings in two input strings, which is referred to as the string kernel in the field of kernel methods. Support vector machine(SVM) is then trained with labeled examples to handle the task of classification. The experiment on hand drawn digital circuit diagrams shows that our system can recognize sketching efficiently and robustly.
{"title":"Sketch recognition via string kernel","authors":"Shizhong Liao, Menghua Duan","doi":"10.1109/ICNC.2012.6234764","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234764","url":null,"abstract":"Sketch recognition is one of the essential step of sketch understanding. Challenge in sketch recognition is the variation and imprecision present in sketch. Free drawing styles of sketching make it difficult to build a robust sketch recognition system. This paper proposes a novel recognition approach that can recognize primitive shapes, as well as combinations of these primitives. The approach is independent of stroke order, number, as well as invariant to size and aspect ratio of sketch. Feature string is used to represent primitives. We defined a similarity measure on these feature strings that counts common substrings in two input strings, which is referred to as the string kernel in the field of kernel methods. Support vector machine(SVM) is then trained with labeled examples to handle the task of classification. The experiment on hand drawn digital circuit diagrams shows that our system can recognize sketching efficiently and robustly.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121245211","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234679
H. Zeng, Lingling Zhou, Linjiang Li Li, Yongqiang Wu
The purpose of this proposes an improved prediction of protein secondary structures based on a multi-mold integrated neural network. A structure of modified artificial neural network based on built a 5-child network integrated multi-mold neural networks in which a child for each network using neural network classification is divided into two-level network is presented. Prediction comprehensive result of protein secondary structure from 5 networks is got. Profile of evolutionary information for protein sequences encoded is taken as an input of a level network. Protein sequences code is added sequence information and prediction of protein is refined by the secondary level network. It is shown that high prediction accuracy of protein secondary structure can be got by an improved multi-mold integrated neural network at 73.1%.
{"title":"An improved prediction of protein secondary structures based on a multi-mold integrated neural network","authors":"H. Zeng, Lingling Zhou, Linjiang Li Li, Yongqiang Wu","doi":"10.1109/ICNC.2012.6234679","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234679","url":null,"abstract":"The purpose of this proposes an improved prediction of protein secondary structures based on a multi-mold integrated neural network. A structure of modified artificial neural network based on built a 5-child network integrated multi-mold neural networks in which a child for each network using neural network classification is divided into two-level network is presented. Prediction comprehensive result of protein secondary structure from 5 networks is got. Profile of evolutionary information for protein sequences encoded is taken as an input of a level network. Protein sequences code is added sequence information and prediction of protein is refined by the secondary level network. It is shown that high prediction accuracy of protein secondary structure can be got by an improved multi-mold integrated neural network at 73.1%.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132774378","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234534
Hongjie Fu
A novel improve algorithm TPDE is proposed in this paper, which combines differential evolution(DE). Each individual contains two states, the attractive state and the repulsive state. In order to refrain from the shortcoming of premature convergence, a two point reversal crossover operator is defined and in the repulsive process each particle is repelled away from some inferior solution in the current tabu list to fly towards some promising areas which can introduce some new information to guide the swarm searching process. DE adjusts the mutation rate F and the crossover rate CR adaptively, taking account of the different distribution of population. TPDE maintains the diversity of population and improves the global convergence ability. It also improves the efficiency and success rate and avoids the premature convergence. Simulation and comparisons based on test-sets of CSPs demonstrate the effectiveness, efficiency and robustness of the proposed algorithm.
{"title":"A novel hybrid alternate two phases differential evolution for binary CSPs","authors":"Hongjie Fu","doi":"10.1109/ICNC.2012.6234534","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234534","url":null,"abstract":"A novel improve algorithm TPDE is proposed in this paper, which combines differential evolution(DE). Each individual contains two states, the attractive state and the repulsive state. In order to refrain from the shortcoming of premature convergence, a two point reversal crossover operator is defined and in the repulsive process each particle is repelled away from some inferior solution in the current tabu list to fly towards some promising areas which can introduce some new information to guide the swarm searching process. DE adjusts the mutation rate F and the crossover rate CR adaptively, taking account of the different distribution of population. TPDE maintains the diversity of population and improves the global convergence ability. It also improves the efficiency and success rate and avoids the premature convergence. Simulation and comparisons based on test-sets of CSPs demonstrate the effectiveness, efficiency and robustness of the proposed algorithm.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133740654","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234507
Zhe Ming Li, Chun Gui Li, S. Lv
This paper describes the use of particle swarm algorithm and k-nearest neighbor method to optimize the process of radial basis function (RBF) network and we use the Denavit-Hartenberg (DH) method to research PUMA560 robotics, the results of the forward kinematics is derived as the RBF network training samples. We use six identical RBF network of twelve-input, single output, to achieve a PUMA560 inverse kinematics calculation. Simulation results show that the results obtained with this method has high accuracy and fast convergence.
{"title":"A method for solving inverse kinematics of PUMA560 manipulator based on PSO-RBF network","authors":"Zhe Ming Li, Chun Gui Li, S. Lv","doi":"10.1109/ICNC.2012.6234507","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234507","url":null,"abstract":"This paper describes the use of particle swarm algorithm and k-nearest neighbor method to optimize the process of radial basis function (RBF) network and we use the Denavit-Hartenberg (DH) method to research PUMA560 robotics, the results of the forward kinematics is derived as the RBF network training samples. We use six identical RBF network of twelve-input, single output, to achieve a PUMA560 inverse kinematics calculation. Simulation results show that the results obtained with this method has high accuracy and fast convergence.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"203 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134205817","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234527
Yanan Wang, Ying-jie Tian
Twin Support Vector Machine (Twin SVM), which is a new binary classifier as an extension of SVMs, was first proposed in 2007 by Jayadeva. Wide attention has been attracted by academic circles for its less computation cost and better generalization ability, and it became a new research priorities gradually. A simple geometric interpretation of Twin SVM is that each hyperplane is closest to the points of its own class and as far as possible from the points of the other class. This method defines two nonparallel hyper-planes by solving two related SVM-type problems. Localized Twin SVM is a classification approach via local information which is based on Twin SVM, and has been proved by experiments having a better performance than conventional Twin SVM. However, the computational cost of the method is so high that it has little practical applications. In this paper we propose a method called Fast Localized Twin SVM, a classifier built so as to be suitable for large data sets, in which the number of Twin SVMs is decreased. In Fast Localized Twin SVM, we first use the training set to compute a set of Localized Twin SVMs, then assign to each local model all the points lying in the central neighborhood of the k training points. The query point depending on its nearest neighbor in the training set can be predicted. From empirical experiments we can show that our approach not only guarantees high generalization ability but also improves the computational cost greatly, especially for large scale data sets.
双支持向量机(Twin Support Vector Machine, Twin SVM)是Jayadeva在2007年提出的一种新的二值分类器，是对支持向量机的扩展。由于其较低的计算成本和较好的泛化能力而受到学术界的广泛关注，并逐渐成为新的研究热点。孪生支持向量机的一个简单的几何解释是，每个超平面最接近自己类的点，并尽可能远离其他类的点。该方法通过求解两个相关的svm型问题来定义两个非平行超平面。局部支持向量机是一种基于双支持向量机的基于局部信息的分类方法，实验证明它比传统的双支持向量机具有更好的性能。然而，该方法的计算成本很高，实际应用很少。在本文中，我们提出了一种称为快速局部双支持向量机的方法，这是一种适合于大数据集的分类器，其中Twin SVM的数量减少。在快速局部孪生支持向量机中，我们首先使用训练集计算一组局部孪生支持向量机，然后将k个训练点的中心邻域内的所有点分配给每个局部模型。查询点依赖于它在训练集中的最近邻居可以被预测。经验实验表明，该方法不仅保证了较高的泛化能力，而且大大提高了计算成本，特别是对于大规模数据集。
{"title":"Fast Localized Twin SVM","authors":"Yanan Wang, Ying-jie Tian","doi":"10.1109/ICNC.2012.6234527","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234527","url":null,"abstract":"Twin Support Vector Machine (Twin SVM), which is a new binary classifier as an extension of SVMs, was first proposed in 2007 by Jayadeva. Wide attention has been attracted by academic circles for its less computation cost and better generalization ability, and it became a new research priorities gradually. A simple geometric interpretation of Twin SVM is that each hyperplane is closest to the points of its own class and as far as possible from the points of the other class. This method defines two nonparallel hyper-planes by solving two related SVM-type problems. Localized Twin SVM is a classification approach via local information which is based on Twin SVM, and has been proved by experiments having a better performance than conventional Twin SVM. However, the computational cost of the method is so high that it has little practical applications. In this paper we propose a method called Fast Localized Twin SVM, a classifier built so as to be suitable for large data sets, in which the number of Twin SVMs is decreased. In Fast Localized Twin SVM, we first use the training set to compute a set of Localized Twin SVMs, then assign to each local model all the points lying in the central neighborhood of the k training points. The query point depending on its nearest neighbor in the training set can be predicted. From empirical experiments we can show that our approach not only guarantees high generalization ability but also improves the computational cost greatly, especially for large scale data sets.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133022164","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234734
Lifeng Zhao, Xiaowan Meng
There are lots of steps and complicated calculation in the existing algorithm for solving the maximum flow,and because of improper selection order of augmented path, we cannot obtain the ideal maximum flow. In order to solve these problems in existing algorithm, this paper make some improvement of the existing algorithms, then puts forward a new improved algorithm for solving the maximum flow problem which make use of divide area and the degree of vertex. And it is verified that the improved algorithm is effective and intuitive through the concrete example, and avoid the labeling process, the entire operation process only needs drawing a diagram to be completed.
{"title":"An improved algorithm for solving maximum flow problem","authors":"Lifeng Zhao, Xiaowan Meng","doi":"10.1109/ICNC.2012.6234734","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234734","url":null,"abstract":"There are lots of steps and complicated calculation in the existing algorithm for solving the maximum flow,and because of improper selection order of augmented path, we cannot obtain the ideal maximum flow. In order to solve these problems in existing algorithm, this paper make some improvement of the existing algorithms, then puts forward a new improved algorithm for solving the maximum flow problem which make use of divide area and the degree of vertex. And it is verified that the improved algorithm is effective and intuitive through the concrete example, and avoid the labeling process, the entire operation process only needs drawing a diagram to be completed.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"211 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134379534","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234510
Jianning Wu
In this paper, we investigated the application of the manifold learning algorithm in gait data analysis for the improvement of the gait classification performance. A manifold learning algorithm such as isometric feature mapping algorithm (ISOMAP) was firstly employed to perform nonlinear feature extraction for initiating the training set, and its effect on a subsequent classification was then tested in combination with learning algorithms such as support vector machines. The gait data including young and elderly participants were analyzed, and the experimental results demonstrated that the generalization performance of ISOMAP-SVM is an evidently improved performance compared to the traditional classifier for recognizing young-elderly gait patterns. Our work suggested that manifold learning algorithm can find the intrinsic low-dimensional manifold embedding in high-dimensional gait data, and obtain the `true' nonlinear gait features associated with human gait function change for improving the gait classification performance. The proposed technique has considerable potential for future clinical applications.
{"title":"Automated recognition of human gait pattern using manifold learning algorithm","authors":"Jianning Wu","doi":"10.1109/ICNC.2012.6234510","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234510","url":null,"abstract":"In this paper, we investigated the application of the manifold learning algorithm in gait data analysis for the improvement of the gait classification performance. A manifold learning algorithm such as isometric feature mapping algorithm (ISOMAP) was firstly employed to perform nonlinear feature extraction for initiating the training set, and its effect on a subsequent classification was then tested in combination with learning algorithms such as support vector machines. The gait data including young and elderly participants were analyzed, and the experimental results demonstrated that the generalization performance of ISOMAP-SVM is an evidently improved performance compared to the traditional classifier for recognizing young-elderly gait patterns. Our work suggested that manifold learning algorithm can find the intrinsic low-dimensional manifold embedding in high-dimensional gait data, and obtain the `true' nonlinear gait features associated with human gait function change for improving the gait classification performance. The proposed technique has considerable potential for future clinical applications.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116233684","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 : 2012-05-29DOI: 10.1109/ICNC.2012.6234596
Bo Xiao, Lijun Guo, Yuanyuan Zhang, Rong-Rrong Zhang
In this paper, we propose a method for human segmentation in videos, extending the recent locally competing 1SVM model. There are only local color distributions to be made use of in the model. To generate a consistent segmentation from complex environments, first, we assume we obtain a bounding box around human by using the human detector. Then we incorporate shape prior information inside the bounding box, which biases the segmentation towards typical human shapes. Finally, we show a substantial improvement over C-1SVM method from our experiment.
{"title":"Human instance segmentation from video using locally competing 1SVMs with shape prior","authors":"Bo Xiao, Lijun Guo, Yuanyuan Zhang, Rong-Rrong Zhang","doi":"10.1109/ICNC.2012.6234596","DOIUrl":"https://doi.org/10.1109/ICNC.2012.6234596","url":null,"abstract":"In this paper, we propose a method for human segmentation in videos, extending the recent locally competing 1SVM model. There are only local color distributions to be made use of in the model. To generate a consistent segmentation from complex environments, first, we assume we obtain a bounding box around human by using the human detector. Then we incorporate shape prior information inside the bounding box, which biases the segmentation towards typical human shapes. Finally, we show a substantial improvement over C-1SVM method from our experiment.","PeriodicalId":404981,"journal":{"name":"2012 8th International Conference on Natural Computation","volume":"235 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116327922","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：