The prediction of Quality of Service (QoS) significantly facilitates the web services selection for QoS based web service recommender systems. One effective method for predicting web services' QoS values is the collaborative filtering (CF) algorithm. However, the existing CF algorithms experience potential scalability issues, as well as the accuracy issues. We present a load- and location-aware collaborative filtering algorithm (LLCF) to improve the prediction accuracy and the scalability. To assess the proposed LLCF, we leverage Amazon Cloud platform where hosts various web services. The experiments are conducted based on selected web services where QoS values are collected. The results show the prediction accuracy is significantly improved by the proposed LLCF. Furthermore, complexity analysis results show that our LLCF can remarkably improve the scalability.
{"title":"LLCF: A Load- and Location-Aware Collaborative Filtering Algorithm to Predict QoS of Web Service","authors":"Chen Li, Xiaochun Zhang, Chengyuan Yu, Xin Shu, Xiaopeng Xu","doi":"10.1109/QRS-C57518.2022.00111","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00111","url":null,"abstract":"The prediction of Quality of Service (QoS) significantly facilitates the web services selection for QoS based web service recommender systems. One effective method for predicting web services' QoS values is the collaborative filtering (CF) algorithm. However, the existing CF algorithms experience potential scalability issues, as well as the accuracy issues. We present a load- and location-aware collaborative filtering algorithm (LLCF) to improve the prediction accuracy and the scalability. To assess the proposed LLCF, we leverage Amazon Cloud platform where hosts various web services. The experiments are conducted based on selected web services where QoS values are collected. The results show the prediction accuracy is significantly improved by the proposed LLCF. Furthermore, complexity analysis results show that our LLCF can remarkably improve the scalability.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128177750","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00029
Taichi Hasegawa, Taiichi Saito, R. Sasaki
In recent years, new types of cyber attacks called targeted attacks have been observed. It targets specific organizations or individuals, while usual large-scale attacks do not focus on specific targets. Organizations have published many Word or PDF files on their websites. These files may provide the starting point for targeted attacks if they include hidden data unintentionally generated in the authoring process. Adhatarao and Lauradoux analyzed hidden data found in the PDF files published by security agencies in many countries and showed that many PDF files potentially leak information like author names, details on the information system and computer architecture. In this study, we analyze hidden data of PDF files published on the website of police agencies in Japan and compare the results with Adhatarao and Lauradoux's. We gathered 110989 PDF files. 56% of gathered PDF files contain personal names, organization names, usernames, or numbers that seem to be IDs within the organizations. 96% of PDF files contain software names.
{"title":"Analyzing Metadata in PDF Files Published by Police Agencies in Japan","authors":"Taichi Hasegawa, Taiichi Saito, R. Sasaki","doi":"10.1109/QRS-C57518.2022.00029","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00029","url":null,"abstract":"In recent years, new types of cyber attacks called targeted attacks have been observed. It targets specific organizations or individuals, while usual large-scale attacks do not focus on specific targets. Organizations have published many Word or PDF files on their websites. These files may provide the starting point for targeted attacks if they include hidden data unintentionally generated in the authoring process. Adhatarao and Lauradoux analyzed hidden data found in the PDF files published by security agencies in many countries and showed that many PDF files potentially leak information like author names, details on the information system and computer architecture. In this study, we analyze hidden data of PDF files published on the website of police agencies in Japan and compare the results with Adhatarao and Lauradoux's. We gathered 110989 PDF files. 56% of gathered PDF files contain personal names, organization names, usernames, or numbers that seem to be IDs within the organizations. 96% of PDF files contain software names.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122001824","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00119
Haoran Sun, Xiaolong Zhu, Conghua Zhou
Video summarization aims to improve the efficiency of large-scale video browsing through producting concise summaries. It has been popular among many scenarios such as video surveillance, video review and data annotation. Traditional video summarization techniques focus on filtration in image features dimension or image semantics dimension. However, such techniques can make a large amount of possible useful information lost, especially for many videos with rich text semantics like interviews, teaching videos, in that only the information relevant to the image dimension will be retained. In order to solve the above problem, this paper considers video summarization as a continuous multi-dimensional decision-making process. Specifically, the summarization model predicts a probability for each frame and its corresponding text, and then we designs reward methods for each of them. Finally, comprehensive summaries in two dimensions, i.e. images and semantics, is generated. This approach is not only unsupervised and does not rely on labels and user interaction, but also decouples the semantic and image summarization models to provide more usable interfaces for subsequent engineering use.
{"title":"Deep Reinforcement Learning for Video Summarization with Semantic Reward","authors":"Haoran Sun, Xiaolong Zhu, Conghua Zhou","doi":"10.1109/QRS-C57518.2022.00119","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00119","url":null,"abstract":"Video summarization aims to improve the efficiency of large-scale video browsing through producting concise summaries. It has been popular among many scenarios such as video surveillance, video review and data annotation. Traditional video summarization techniques focus on filtration in image features dimension or image semantics dimension. However, such techniques can make a large amount of possible useful information lost, especially for many videos with rich text semantics like interviews, teaching videos, in that only the information relevant to the image dimension will be retained. In order to solve the above problem, this paper considers video summarization as a continuous multi-dimensional decision-making process. Specifically, the summarization model predicts a probability for each frame and its corresponding text, and then we designs reward methods for each of them. Finally, comprehensive summaries in two dimensions, i.e. images and semantics, is generated. This approach is not only unsupervised and does not rely on labels and user interaction, but also decouples the semantic and image summarization models to provide more usable interfaces for subsequent engineering use.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126675077","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00104
Changjian Li, Dongcheng Li, Man Zhao, Hui Li
This paper builds a convolutional neural network model based on Xception; we delete the fully connected layer in the traditional convolutional neural network model and use four depthwise separable convolutions to replace the convolution layer in convolutional neural network; we use batch normalization to process the output data after each convolution operation, and use the ReLU activation function to add nonlinear factors to the output data, and finally use the SoftMax function for final result classification. Our model achieved an accuracy rate of 73% on the FER2013 dataset, which is a particular improvement compared to the original model Xception. We design and implement a facial recognition system that can be used for static images and real-time recognition, which can quickly and accurately recognize authentic facial expressions.
{"title":"A Light-Weight Convolutional Neural Network for Facial Expression Recognition using Mini-Xception Neural Networks","authors":"Changjian Li, Dongcheng Li, Man Zhao, Hui Li","doi":"10.1109/QRS-C57518.2022.00104","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00104","url":null,"abstract":"This paper builds a convolutional neural network model based on Xception; we delete the fully connected layer in the traditional convolutional neural network model and use four depthwise separable convolutions to replace the convolution layer in convolutional neural network; we use batch normalization to process the output data after each convolution operation, and use the ReLU activation function to add nonlinear factors to the output data, and finally use the SoftMax function for final result classification. Our model achieved an accuracy rate of 73% on the FER2013 dataset, which is a particular improvement compared to the original model Xception. We design and implement a facial recognition system that can be used for static images and real-time recognition, which can quickly and accurately recognize authentic facial expressions.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130089715","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00034
Chenxi He, Wenhong Liu, Shuang Zhao, Jiawei Liu, Yang Yang
Due to the rapid development of deep neural networks, in recent years, machine translation software has been widely adopted in people's daily lives, such as communicating with foreigners or understanding political news from the neighbouring countries, and it is embedded in daily applications such as Twitter and WeChat. The neural machine translation (NMT) model is the core of machine translation software, and it is very challenging to test it as a deep neural network model due to the Inexplicability of neural networks and the complexity of model output. In this paper, we introduce three latest machine translation testing methods and provide a preliminary analysis of their effects.
{"title":"An Empirical Study towards Characterizing Neural Machine Translation Testing Methods","authors":"Chenxi He, Wenhong Liu, Shuang Zhao, Jiawei Liu, Yang Yang","doi":"10.1109/QRS-C57518.2022.00034","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00034","url":null,"abstract":"Due to the rapid development of deep neural networks, in recent years, machine translation software has been widely adopted in people's daily lives, such as communicating with foreigners or understanding political news from the neighbouring countries, and it is embedded in daily applications such as Twitter and WeChat. The neural machine translation (NMT) model is the core of machine translation software, and it is very challenging to test it as a deep neural network model due to the Inexplicability of neural networks and the complexity of model output. In this paper, we introduce three latest machine translation testing methods and provide a preliminary analysis of their effects.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129192280","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00085
Huan Jin, Hongyan Wan, Ziruo Li, Wenxuan Wang
User comments are one of the main ways for IT companies to obtain software evolution requirements. There are two major methods used to classify the software requirements: the traditional user requirements mining method and the user comments requirements mining. The advantage of traditional user requirements mining is that it can communicate with users face to face, but it is time consuming and the results may not be accurate. Therefore, in this paper, we use the user comments requirements mining method to compare the labeling effect of classification methods on the data set of 19,673 comments. The experimental results show that the combination of TF-IDF and logistic regression (LR) works best on the labeled dataset. This experiment combined with word cloud map has excellent effect on obtaining user requirements.
{"title":"An Empirical Study on Software Requirements Classification Method based on Mobile App User Comments","authors":"Huan Jin, Hongyan Wan, Ziruo Li, Wenxuan Wang","doi":"10.1109/QRS-C57518.2022.00085","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00085","url":null,"abstract":"User comments are one of the main ways for IT companies to obtain software evolution requirements. There are two major methods used to classify the software requirements: the traditional user requirements mining method and the user comments requirements mining. The advantage of traditional user requirements mining is that it can communicate with users face to face, but it is time consuming and the results may not be accurate. Therefore, in this paper, we use the user comments requirements mining method to compare the labeling effect of classification methods on the data set of 19,673 comments. The experimental results show that the combination of TF-IDF and logistic regression (LR) works best on the labeled dataset. This experiment combined with word cloud map has excellent effect on obtaining user requirements.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131808037","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00117
Chuan Zhao, Huilei Cao
We consider a destination port logistics service provider (DPLSP), which wants to improve its service quality by reducing risk of delivery time delay. This paper diagnoses potential risk factors that estimate the performances of the DPLSP who provides services only after the arrival of freight, with the intention of reducing supply chain risk and improve supply chain performance through creative computing approach. Self-organizing feature map (SOFM) computing is a type of artificial neural network based on an unsupervised learning algorithm. We propose the approach of SOFM computing for the purpose of clustering risk data of DPLSPs from a less subjective perspective and then rank the cluster results into different levels based on the total risk value of each cluster. Numerical studies to test the effectiveness of this model would be carried out using air import logistics lead-time reports from a large DPLSP. The results illustrate that the proposed approach could successfully cluster and rank the risk data according to their values.
{"title":"Risk Evaluation of the Destination Port Logistics based on Self-Organizing Map Computing","authors":"Chuan Zhao, Huilei Cao","doi":"10.1109/QRS-C57518.2022.00117","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00117","url":null,"abstract":"We consider a destination port logistics service provider (DPLSP), which wants to improve its service quality by reducing risk of delivery time delay. This paper diagnoses potential risk factors that estimate the performances of the DPLSP who provides services only after the arrival of freight, with the intention of reducing supply chain risk and improve supply chain performance through creative computing approach. Self-organizing feature map (SOFM) computing is a type of artificial neural network based on an unsupervised learning algorithm. We propose the approach of SOFM computing for the purpose of clustering risk data of DPLSPs from a less subjective perspective and then rank the cluster results into different levels based on the total risk value of each cluster. Numerical studies to test the effectiveness of this model would be carried out using air import logistics lead-time reports from a large DPLSP. The results illustrate that the proposed approach could successfully cluster and rank the risk data according to their values.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129265858","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00112
Lingfei Ma, Liming Nie, Chenxi Mao, Yaowen Zheng, Y. Liu
In this paper, we propose an analytical model that can analyze the impact of emergencies on open source software (OSS) development. As the core of this model, a metric system is used to comprehensively describe the OSS development process, which includes three dimensions: team activity, development activity, and development risk, with a total of 30 metrics. To demonstrate the effectiveness of the model, we construct an empirical study analyzing the impact of COVID-19 on OSS development. This study is based on the development process events between January 2019 and April 2022 belonging to 50 selected open source projects on GitHub. The results show that more than 72.4% of projects were negatively impacted following the COVID-19 outbreak. Interestingly, we observe that variants of covide-19 did not exacerbate its impact on software development. On the contrary, some project development activities have obviously resumed, indicating that the development team has adapted and gradually got rid of the impact of the epidemic.
{"title":"An Empirical Study of the Impact of COVID-19 on OSS Development","authors":"Lingfei Ma, Liming Nie, Chenxi Mao, Yaowen Zheng, Y. Liu","doi":"10.1109/QRS-C57518.2022.00112","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00112","url":null,"abstract":"In this paper, we propose an analytical model that can analyze the impact of emergencies on open source software (OSS) development. As the core of this model, a metric system is used to comprehensively describe the OSS development process, which includes three dimensions: team activity, development activity, and development risk, with a total of 30 metrics. To demonstrate the effectiveness of the model, we construct an empirical study analyzing the impact of COVID-19 on OSS development. This study is based on the development process events between January 2019 and April 2022 belonging to 50 selected open source projects on GitHub. The results show that more than 72.4% of projects were negatively impacted following the COVID-19 outbreak. Interestingly, we observe that variants of covide-19 did not exacerbate its impact on software development. On the contrary, some project development activities have obviously resumed, indicating that the development team has adapted and gradually got rid of the impact of the epidemic.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115921747","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00092
M. Tuo, Xiaoqiang Zhao, Bo Shen, Wen-Ling Wu
In this paper, a formal method for real-time verification of Cyber-Physical Systems(CPS) is proposed. Firstly, the CPS behavior model is established by using temporal automata, and then the real-time verification of the system is performed. Based on this method, the real-time task scheduling of the control software of an intelligent car is analyzed. The experimental results show the effectiveness of this method.
{"title":"Modeling and Real-Time Verification for CPS based on Time Automata","authors":"M. Tuo, Xiaoqiang Zhao, Bo Shen, Wen-Ling Wu","doi":"10.1109/QRS-C57518.2022.00092","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00092","url":null,"abstract":"In this paper, a formal method for real-time verification of Cyber-Physical Systems(CPS) is proposed. Firstly, the CPS behavior model is established by using temporal automata, and then the real-time verification of the system is performed. Based on this method, the real-time task scheduling of the control software of an intelligent car is analyzed. The experimental results show the effectiveness of this method.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116966530","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 : 2022-12-01DOI: 10.1109/QRS-C57518.2022.00078
Xiao Chen, Junhua Wu
Developers tend to search and reuse code snippets from large-scale corpora while implementing some of the features that existed in development. This will improve the efficiency of development. Code search is to search for semantically relevant code snippets based on a given natural language query. In existing methods, the semantic similarity between code and query is quantified as their distance in the shared vector space. To improve the vector space and map the code vector and query vector into a shared vector space so that the semantically similar code-query pairs are close to each other, we propose a code search method with multimodal representations. It can better enhance the semantic relationship between code snippets and queries. Experiments on Java datasets show that the multimodal representation model MulCS improves the quality of code search. MulCS outperforms several existing advanced models in several performance metrics.
{"title":"Code Search Method Based on Multimodal Representation","authors":"Xiao Chen, Junhua Wu","doi":"10.1109/QRS-C57518.2022.00078","DOIUrl":"https://doi.org/10.1109/QRS-C57518.2022.00078","url":null,"abstract":"Developers tend to search and reuse code snippets from large-scale corpora while implementing some of the features that existed in development. This will improve the efficiency of development. Code search is to search for semantically relevant code snippets based on a given natural language query. In existing methods, the semantic similarity between code and query is quantified as their distance in the shared vector space. To improve the vector space and map the code vector and query vector into a shared vector space so that the semantically similar code-query pairs are close to each other, we propose a code search method with multimodal representations. It can better enhance the semantic relationship between code snippets and queries. Experiments on Java datasets show that the multimodal representation model MulCS improves the quality of code search. MulCS outperforms several existing advanced models in several performance metrics.","PeriodicalId":183728,"journal":{"name":"2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114565754","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：