Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152777
Yin Xie, Yigui Luo, Haihong She, Zhaohong Xiang
In past work, deep learning researchers always designed hyperparameters such as model structure and learning rate first and then used the training set to train the weights in this model. While unrestricted model structure design leads to massive neuron redundancy in neural network models. By pruning these redundant neurons, not only can the storage be compressed effectively, but also the operation can be accelerated. In this paper, we propose a method to utilize the training set to prune the model structure during training: 1) train the initialized model and bring it to basic convergence; 2) feed the entire training set into the model and calculate the activations of neurons in each layer; 3) calculate the threshold for neuron pruning in each layer according to the pruning ratio, delete neurons whose activation value is lower than the threshold, and correspondingly delete the weights of the upper and lower layers; 4) further train the pruned model so that it eventually converges. This method of deleting redundant neurons not only greatly deletes the parameters in the model but also achieves model acceleration. We applied this method to some mainstream neural network models: VGGNet and ResNet, and achieved good results.
{"title":"Neural Network Model Pruning without Additional Computation and Structure Requirements","authors":"Yin Xie, Yigui Luo, Haihong She, Zhaohong Xiang","doi":"10.1109/CSCWD57460.2023.10152777","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152777","url":null,"abstract":"In past work, deep learning researchers always designed hyperparameters such as model structure and learning rate first and then used the training set to train the weights in this model. While unrestricted model structure design leads to massive neuron redundancy in neural network models. By pruning these redundant neurons, not only can the storage be compressed effectively, but also the operation can be accelerated. In this paper, we propose a method to utilize the training set to prune the model structure during training: 1) train the initialized model and bring it to basic convergence; 2) feed the entire training set into the model and calculate the activations of neurons in each layer; 3) calculate the threshold for neuron pruning in each layer according to the pruning ratio, delete neurons whose activation value is lower than the threshold, and correspondingly delete the weights of the upper and lower layers; 4) further train the pruned model so that it eventually converges. This method of deleting redundant neurons not only greatly deletes the parameters in the model but also achieves model acceleration. We applied this method to some mainstream neural network models: VGGNet and ResNet, and achieved good results.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"17 1","pages":"1734-1740"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89679863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152611
Ya Zhang, Xi Lin, Jun Wu, Bei Pei, Yunyun Han
The flexibility and low cost of unmanned aerial vehicles (UAVs) offer great potential for them in areas such as disaster relief, energy line inspection, and traffic monitoring. Multiple UAVs form an airborne UAV network to share geo-tagged observation data for better collaborative missions. Blockchain can solve the security threats caused by the environment’s untrustworthiness and the UAV networks’ openness. However, the key to sharing data in blockchain-assisted UAV networks is identifying and understanding observational data and providing authentication query services in free boundary spatial. This paper proposes a blockchain-assisted UAV network data-sharing framework based on Non-Fungible Token (NFT). First, we design a marking and describing data method based on NFT to help geo-tagged data be effectively understood. Moreover, we propose a free-boundary spatial index tree to manage data and provide efficient queries. Furthermore, combined with the consensus mechanism and the blockchain transaction tree, the proposed sharing framework can provide query results authentication. Compared with the existing schemes, analysis and experiments demonstrate that our scheme could support the arbitrary expansion of UAV flight range and the random distribution of observation data in space, save at least 22% of storage overhead and reduce more than 36% of query time overhead.
{"title":"Blockchain-Assisted UAV Data Free-Boundary Spatial Querying and Authenticated Sharing","authors":"Ya Zhang, Xi Lin, Jun Wu, Bei Pei, Yunyun Han","doi":"10.1109/CSCWD57460.2023.10152611","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152611","url":null,"abstract":"The flexibility and low cost of unmanned aerial vehicles (UAVs) offer great potential for them in areas such as disaster relief, energy line inspection, and traffic monitoring. Multiple UAVs form an airborne UAV network to share geo-tagged observation data for better collaborative missions. Blockchain can solve the security threats caused by the environment’s untrustworthiness and the UAV networks’ openness. However, the key to sharing data in blockchain-assisted UAV networks is identifying and understanding observational data and providing authentication query services in free boundary spatial. This paper proposes a blockchain-assisted UAV network data-sharing framework based on Non-Fungible Token (NFT). First, we design a marking and describing data method based on NFT to help geo-tagged data be effectively understood. Moreover, we propose a free-boundary spatial index tree to manage data and provide efficient queries. Furthermore, combined with the consensus mechanism and the blockchain transaction tree, the proposed sharing framework can provide query results authentication. Compared with the existing schemes, analysis and experiments demonstrate that our scheme could support the arbitrary expansion of UAV flight range and the random distribution of observation data in space, save at least 22% of storage overhead and reduce more than 36% of query time overhead.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"169 1","pages":"565-570"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78818936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152817
Jiayan Xiang, Wanjun Chen, Yang Wang, Bowen Liang, Zihao Liu, Guosheng Kang
With the development of Mashup technique, the number of Web APIs released on the Web continues to grow year by year. However, it is a challenging issue to find and select the desirable Web APIs among the large amount of Web APIs. Consequently, interactive Web API recommendation is used to alleviate the difficulty of service selection, when users or developers try to invoke Web APIs for solving their business requirements or software development requirements. Currently, there are several collaborative filtering based approaches proposed for Web API recommendation, while their recommendation performance is limited on both optimality and scalability. This paper proposes a light neural graph collaborative filtering based Web API recommendation approach, named LNGCF. Specifically, LNGCF learns user and item embeddings by linearly propagating them on the user-item interaction graph, and uses the weighted summation of the embeddings learned at all layers as the final embedding. Such simple, linear, and neat model is much easier to implement and train. A set of experiments are conducted on a real-world dataset. Experimental results demonstrate the substantial improvements on both optimality and scalability over the baselines.
{"title":"Interactive Web API Recommendation for Mashup Development based on Light Neural Graph Collaborative Filtering","authors":"Jiayan Xiang, Wanjun Chen, Yang Wang, Bowen Liang, Zihao Liu, Guosheng Kang","doi":"10.1109/CSCWD57460.2023.10152817","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152817","url":null,"abstract":"With the development of Mashup technique, the number of Web APIs released on the Web continues to grow year by year. However, it is a challenging issue to find and select the desirable Web APIs among the large amount of Web APIs. Consequently, interactive Web API recommendation is used to alleviate the difficulty of service selection, when users or developers try to invoke Web APIs for solving their business requirements or software development requirements. Currently, there are several collaborative filtering based approaches proposed for Web API recommendation, while their recommendation performance is limited on both optimality and scalability. This paper proposes a light neural graph collaborative filtering based Web API recommendation approach, named LNGCF. Specifically, LNGCF learns user and item embeddings by linearly propagating them on the user-item interaction graph, and uses the weighted summation of the embeddings learned at all layers as the final embedding. Such simple, linear, and neat model is much easier to implement and train. A set of experiments are conducted on a real-world dataset. Experimental results demonstrate the substantial improvements on both optimality and scalability over the baselines.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"31 1","pages":"1926-1931"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79398217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/cscwd57460.2023.10152747
{"title":"Keynote 4 - The Good, The Bad, and The Ethical: Exploring Artificial Intelligence in Automatic Decision-Making","authors":"","doi":"10.1109/cscwd57460.2023.10152747","DOIUrl":"https://doi.org/10.1109/cscwd57460.2023.10152747","url":null,"abstract":"","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"37 1","pages":""},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74658653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152565
Ramon Chaves, C. Motta, António Correia, Jano de Souza, D. Schneider
This paper explores the challenges and strategies used to design, build and evaluate a collective intelligence (CI) model to support discussions about cities. Through an autoethnography of ideas and discussion practices in both online and offline contexts, this work explores the tensions between the chosen methodological approaches, including design science research (DSR) and participatory action research (PAR). Moreover, we also examine the challenges and pitfalls observed during the practical conduction of this research when involving participants in the process of empirically evaluating the proposed model. Finally, aspects related to the autoethnographic process itself as a reflective method are discussed alongside the consequences of desiring and seeking participation in the research process.
{"title":"Tensions in design and participation processes: An ethnographic approach to the design, building and evaluation of a collective intelligence model","authors":"Ramon Chaves, C. Motta, António Correia, Jano de Souza, D. Schneider","doi":"10.1109/CSCWD57460.2023.10152565","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152565","url":null,"abstract":"This paper explores the challenges and strategies used to design, build and evaluate a collective intelligence (CI) model to support discussions about cities. Through an autoethnography of ideas and discussion practices in both online and offline contexts, this work explores the tensions between the chosen methodological approaches, including design science research (DSR) and participatory action research (PAR). Moreover, we also examine the challenges and pitfalls observed during the practical conduction of this research when involving participants in the process of empirically evaluating the proposed model. Finally, aspects related to the autoethnographic process itself as a reflective method are discussed alongside the consequences of desiring and seeking participation in the research process.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"304 1","pages":"462-467"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76557585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152852
Ning Zhang, Tianyue Qiu, Liuliu Du-Ikonen, Xiaojie Lin, Qichao Ye, Jiaying Chen
A smart energy platform for the large-space stadium based on Internet of Things (IoT) is proposed. The platform could realize the safe and stable operation of the energy system in various scenarios and promote low-carbon, efficient and sustainable development. In this paper, the smart energy platform is constructed based on IoT device data collection, time series data storage, front-end smart energy platform, and back-end optimization modules. The architecture and framework of the platform are explained in details. This study takes a large-space building in Hangzhou as an example to explain how the smart energy platform works in the real site. The real-time monitoring map of carbon emissions module and load prediction module of air-conditioning system are presented. The developed smart energy platform based on IoT could support the digital twin-based operation management of various types of low-carbon buildings in the future.
{"title":"Smart Energy Platform for Large-space Stadium Construction Based on Internet of Things","authors":"Ning Zhang, Tianyue Qiu, Liuliu Du-Ikonen, Xiaojie Lin, Qichao Ye, Jiaying Chen","doi":"10.1109/CSCWD57460.2023.10152852","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152852","url":null,"abstract":"A smart energy platform for the large-space stadium based on Internet of Things (IoT) is proposed. The platform could realize the safe and stable operation of the energy system in various scenarios and promote low-carbon, efficient and sustainable development. In this paper, the smart energy platform is constructed based on IoT device data collection, time series data storage, front-end smart energy platform, and back-end optimization modules. The architecture and framework of the platform are explained in details. This study takes a large-space building in Hangzhou as an example to explain how the smart energy platform works in the real site. The real-time monitoring map of carbon emissions module and load prediction module of air-conditioning system are presented. The developed smart energy platform based on IoT could support the digital twin-based operation management of various types of low-carbon buildings in the future.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"119 1","pages":"762-765"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76689294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152597
Xiaolong Xu, Yihao Dong, Jian Song
At present, deep learning technologies have been widely used in the field of natural language process, such as text summarization. In CQA, the answer summary could help users get a complete answer quickly. There are still some problems with the current answer summary scheme, such as semantic inconsistency, repetition of words, etc. In order to solve this, we propose a novel scheme Answer Summarization based on Multi-layer Attention Scheme (ASMAM). Based on the traditional Seq2Seq, we introduce self-attention and multi-head attention scheme respectively during sentence and text encoding, which could improve text representation ability of the model. In order to solve "long distance dependence" of RNN and too many parameters of LSTM, we all use GRU as the neuron at the encoder and decoder sides. Experiments over the Yahoo! Answers dataset demonstrate that the coherence and fluency of the generated summary are all superior to the benchmark model in ROUGE evaluation system.
{"title":"An Answer Summarization Scheme Based on Multilayer Attention Model","authors":"Xiaolong Xu, Yihao Dong, Jian Song","doi":"10.1109/CSCWD57460.2023.10152597","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152597","url":null,"abstract":"At present, deep learning technologies have been widely used in the field of natural language process, such as text summarization. In CQA, the answer summary could help users get a complete answer quickly. There are still some problems with the current answer summary scheme, such as semantic inconsistency, repetition of words, etc. In order to solve this, we propose a novel scheme Answer Summarization based on Multi-layer Attention Scheme (ASMAM). Based on the traditional Seq2Seq, we introduce self-attention and multi-head attention scheme respectively during sentence and text encoding, which could improve text representation ability of the model. In order to solve \"long distance dependence\" of RNN and too many parameters of LSTM, we all use GRU as the neuron at the encoder and decoder sides. Experiments over the Yahoo! Answers dataset demonstrate that the coherence and fluency of the generated summary are all superior to the benchmark model in ROUGE evaluation system.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"7 1 1","pages":"143-148"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77619241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-24DOI: 10.1109/CSCWD57460.2023.10152748
Han Zhang, Lubin Lin, Guipeng Zhang, Zhenguo Yang, Wenyin Liu
With the spread of information on the Internet and the explosive growth of intellectual property information, the traditional intellectual property management model relying on third-party institutions cannot meet the demand for intellectual property protection, which has a cumbersome process, low efficiency, and insufficient evidence of rights protection. To address the issues of information falsification and leakage, we present an anonymous and traceable intellectual property management system based on blockchain, namely ATIPM, which employs the non-interactive zero knowledge proof to realize user unlinkability and anonymous transactions to protect the users’ intellectual property information. To avoid a single point of accountability, the ATIPM introduces a threshold ramp secret sharing scheme to achieve the traceability of intellectual property for all users and greatly improve the users’ privacy security and autonomy by preventing information leakage from malicious third-party institutions. Furthermore, the ATIPM can improve the management efficiency of intellectual property by utilizing smart contracts to realize efficient retrieval and verification of intellectual property. The evaluation results demonstrate the effectiveness of our proposed system.
{"title":"ATIPM: A Blockchain-Based Anonymous and Traceable Intellectual Property Management Scheme","authors":"Han Zhang, Lubin Lin, Guipeng Zhang, Zhenguo Yang, Wenyin Liu","doi":"10.1109/CSCWD57460.2023.10152748","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152748","url":null,"abstract":"With the spread of information on the Internet and the explosive growth of intellectual property information, the traditional intellectual property management model relying on third-party institutions cannot meet the demand for intellectual property protection, which has a cumbersome process, low efficiency, and insufficient evidence of rights protection. To address the issues of information falsification and leakage, we present an anonymous and traceable intellectual property management system based on blockchain, namely ATIPM, which employs the non-interactive zero knowledge proof to realize user unlinkability and anonymous transactions to protect the users’ intellectual property information. To avoid a single point of accountability, the ATIPM introduces a threshold ramp secret sharing scheme to achieve the traceability of intellectual property for all users and greatly improve the users’ privacy security and autonomy by preventing information leakage from malicious third-party institutions. Furthermore, the ATIPM can improve the management efficiency of intellectual property by utilizing smart contracts to realize efficient retrieval and verification of intellectual property. The evaluation results demonstrate the effectiveness of our proposed system.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"1 1","pages":"1080-1085"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77622604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the production process of automobiles, parts procurement is invariably a crucial step. In order to find an optimal decision, it is a challenge to match parts to suppliers for the limited financial and material capabilities of every supplier. This paper formalized the problem by Group Multirole Assignment (GMRA). Meanwhile, the success of this assignment process depends on the choice of the agent evaluation method. It depends on the industrial chain data, which can acquire feature indexes of parts from previous purchase records. Furthermore, comprehensive evaluation of parts procurement bases on multiple factors. Thus, it is difficult to reflect different quantifications using the multifactorial parameter semantics. Therefore, we propose a new method of Fuzzy Hierarchy Comprehensive Evaluation (FHCE), using membership grades of the fuzzy theory to differentiate the parameter and the weight, which can use objective quantitative analysis to optimize procurement plan. After that, based on GMRA, decision makers are able to maximize the resource utilization ratio to determine optimized solutions when funds or part types are limited. Simulation experiments indicate that the proposed method is efficient and feasible, which is verified practicable.
{"title":"Industrial Chain Data Evaluation in Automobile Parts Procurement via Group Multirole Assignment","authors":"Ziqi Xiong, Haibin Zhu, Dongning Liu, Jianhui Xian","doi":"10.1109/CSCWD57460.2023.10152642","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152642","url":null,"abstract":"In the production process of automobiles, parts procurement is invariably a crucial step. In order to find an optimal decision, it is a challenge to match parts to suppliers for the limited financial and material capabilities of every supplier. This paper formalized the problem by Group Multirole Assignment (GMRA). Meanwhile, the success of this assignment process depends on the choice of the agent evaluation method. It depends on the industrial chain data, which can acquire feature indexes of parts from previous purchase records. Furthermore, comprehensive evaluation of parts procurement bases on multiple factors. Thus, it is difficult to reflect different quantifications using the multifactorial parameter semantics. Therefore, we propose a new method of Fuzzy Hierarchy Comprehensive Evaluation (FHCE), using membership grades of the fuzzy theory to differentiate the parameter and the weight, which can use objective quantitative analysis to optimize procurement plan. After that, based on GMRA, decision makers are able to maximize the resource utilization ratio to determine optimized solutions when funds or part types are limited. Simulation experiments indicate that the proposed method is efficient and feasible, which is verified practicable.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"62 1","pages":"1049-1054"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77912880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The rapid development of pornographic and gambling websites, fueled by the widespread abuse of information technology, has become a growing concern. They pose a serious threat to the physical and mental health of children and can also endanger personal property. Therefore, it is necessary to detect them. However, pornographic and gambling websites become more and more tricky, which shows fake-normal to evade censorship and challenges traditional content-based detection methods. Therefore, it is essential to rely on information about relationships between websites.We propose HMAN, one Multi-Attention Heterogeneous Graph Neural Network (HGNN) model to detect pornographic and gambling websites by integrating content features and structural information, even if they present fake-normal. By one multi-attention mechanism consisting of explicit weight, self-attention and attention mechanism, content features can be selectively utilized with the assistance of structural information. The experimental results show that our method achieves the best 95.1% Macro-Avg-F1 and outperforms all baselines. We also illustrate that all extracted metapaths do contribute to the detection, where the hyperlink, title/meta terms and IP address are relatively important.
{"title":"Detecting Fake-Normal Pornographic and Gambling Websites through one Multi-Attention HGNN","authors":"Xiaoqing Ma, Chao Zheng, Zhao Li, Jiang Yin, Qingyun Liu, Xunxun Chen","doi":"10.1109/CSCWD57460.2023.10152775","DOIUrl":"https://doi.org/10.1109/CSCWD57460.2023.10152775","url":null,"abstract":"The rapid development of pornographic and gambling websites, fueled by the widespread abuse of information technology, has become a growing concern. They pose a serious threat to the physical and mental health of children and can also endanger personal property. Therefore, it is necessary to detect them. However, pornographic and gambling websites become more and more tricky, which shows fake-normal to evade censorship and challenges traditional content-based detection methods. Therefore, it is essential to rely on information about relationships between websites.We propose HMAN, one Multi-Attention Heterogeneous Graph Neural Network (HGNN) model to detect pornographic and gambling websites by integrating content features and structural information, even if they present fake-normal. By one multi-attention mechanism consisting of explicit weight, self-attention and attention mechanism, content features can be selectively utilized with the assistance of structural information. The experimental results show that our method achieves the best 95.1% Macro-Avg-F1 and outperforms all baselines. We also illustrate that all extracted metapaths do contribute to the detection, where the hyperlink, title/meta terms and IP address are relatively important.","PeriodicalId":51008,"journal":{"name":"Computer Supported Cooperative Work-The Journal of Collaborative Computing","volume":"1 1","pages":"1741-1747"},"PeriodicalIF":2.4,"publicationDate":"2023-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83085291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","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：