Pub Date : 2022-08-19DOI: 10.1109/CCET55412.2022.9906329
Yongmei Zhang, Bin Dai, Minghui Dong, Hao Chen, Mengyang Zhou
Lung cancer is the malignant tumor with the highest morbidity and mortality, and it is a great threat to human health. With the increasing refinement of lung cancer images, it provides a lot of useful information for the analysis and identification of lung cancer, and an important help to assist doctors in making accurate diagnosis. A considerable part of lung cancer manifests as nodules in the early stage. Pulmonary nodules are round or irregular lesions in the lungs, about 34% are lung cancers, and the rest are benign lesions. Therefore, the detection of pulmonary nodules is very important for the detection of early lung cancer. In this paper, some Computed Tomography (CT) images of the Lung Image Database Consortium (LIDC) dataset are adopted as training and testing data, data preprocessing is completed by intercepting pixels, normalization and other methods, data enhancement is realized such as rotation and scaling methods, and the pulmonary nodule sample library is expanded. Utilizing the constructed lung nodule sample library, train the Convolutional Neural Network (CNN) model, complete the detection and segmentation of pulmonary nodules, and exact the regions of pulmonary nodules. The size and regularity features of pulmonary nodules are extracted, and lung cancer recognition is realized according to the size and shape of pulmonary nodules. The experiment results show the lung cancer detection and identification method based on convolutional neural network with morphological features has higher accuracy.
{"title":"A Lung Cancer Detection and Recognition Method Combining Convolutional Neural Network and Morphological Features","authors":"Yongmei Zhang, Bin Dai, Minghui Dong, Hao Chen, Mengyang Zhou","doi":"10.1109/CCET55412.2022.9906329","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906329","url":null,"abstract":"Lung cancer is the malignant tumor with the highest morbidity and mortality, and it is a great threat to human health. With the increasing refinement of lung cancer images, it provides a lot of useful information for the analysis and identification of lung cancer, and an important help to assist doctors in making accurate diagnosis. A considerable part of lung cancer manifests as nodules in the early stage. Pulmonary nodules are round or irregular lesions in the lungs, about 34% are lung cancers, and the rest are benign lesions. Therefore, the detection of pulmonary nodules is very important for the detection of early lung cancer. In this paper, some Computed Tomography (CT) images of the Lung Image Database Consortium (LIDC) dataset are adopted as training and testing data, data preprocessing is completed by intercepting pixels, normalization and other methods, data enhancement is realized such as rotation and scaling methods, and the pulmonary nodule sample library is expanded. Utilizing the constructed lung nodule sample library, train the Convolutional Neural Network (CNN) model, complete the detection and segmentation of pulmonary nodules, and exact the regions of pulmonary nodules. The size and regularity features of pulmonary nodules are extracted, and lung cancer recognition is realized according to the size and shape of pulmonary nodules. The experiment results show the lung cancer detection and identification method based on convolutional neural network with morphological features has higher accuracy.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125276416","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}
Masks will invalidate the original face recognition algorithm model and make the computer unable to recognize faces. In addition, there are many types of masks, and the degree of occlusion is different, which increases the difficulty of face recognition. This paper combines the traditional feature extraction method with deep learning, and proposes a face recognition method with masks based on low-level features and deep residual network. The method of face segmentation based on feature points is used to extract the local features of the face, using the Holistically-nested Edge Detection (HED) algorithm to extract the overall contour features of the face, fusion of local features, overall contour features and pre-processed images into a deep residual network model, realize face recognition with masks, and evaluate the face recognition method with accuracy. The experiment results show this method improves the recognition accuracy compared with Principal Component Analysis (PCA) and convolutional neural network (CNN).
{"title":"A Mask-Wearing Face Recognition Method Based on Low-Level Features and Deep Residual Networks","authors":"Yongmei Zhang, Chenyang Sun, Mengyang Zhou, Haoxing Chen, Minghui Dong","doi":"10.1109/CCET55412.2022.9906328","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906328","url":null,"abstract":"Masks will invalidate the original face recognition algorithm model and make the computer unable to recognize faces. In addition, there are many types of masks, and the degree of occlusion is different, which increases the difficulty of face recognition. This paper combines the traditional feature extraction method with deep learning, and proposes a face recognition method with masks based on low-level features and deep residual network. The method of face segmentation based on feature points is used to extract the local features of the face, using the Holistically-nested Edge Detection (HED) algorithm to extract the overall contour features of the face, fusion of local features, overall contour features and pre-processed images into a deep residual network model, realize face recognition with masks, and evaluate the face recognition method with accuracy. The experiment results show this method improves the recognition accuracy compared with Principal Component Analysis (PCA) and convolutional neural network (CNN).","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131047472","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906350
Xiangjv Sun, Wenhao Zhan, Y. Li
Today, operation and maintenance security has become a key factor for the stable development of enterprises. The flexible alarm information communication scheduling structure and intelligent fault alarm method are of great significance to improve the accuracy of alarm judgment, which can greatly realize the safe production of power enterprises and effectively ensure their steady and healthy development. Therefore, this paper proposes a modular structure. The alarm information communication scheduling architecture of the system uses unsupervised machine learning algorithms to predict fault information, analyze the correlation between faults, and at the same time achieve a high degree of coordination between modules, thereby providing an efficient and stable alarm method.
{"title":"Fault Alarm Communication Scheduling Architecture and Method for Power System","authors":"Xiangjv Sun, Wenhao Zhan, Y. Li","doi":"10.1109/CCET55412.2022.9906350","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906350","url":null,"abstract":"Today, operation and maintenance security has become a key factor for the stable development of enterprises. The flexible alarm information communication scheduling structure and intelligent fault alarm method are of great significance to improve the accuracy of alarm judgment, which can greatly realize the safe production of power enterprises and effectively ensure their steady and healthy development. Therefore, this paper proposes a modular structure. The alarm information communication scheduling architecture of the system uses unsupervised machine learning algorithms to predict fault information, analyze the correlation between faults, and at the same time achieve a high degree of coordination between modules, thereby providing an efficient and stable alarm method.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"114 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132567219","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 : 2022-08-19DOI: 10.1109/ccet55412.2022.9906327
{"title":"CCET 2022 Cover Page","authors":"","doi":"10.1109/ccet55412.2022.9906327","DOIUrl":"https://doi.org/10.1109/ccet55412.2022.9906327","url":null,"abstract":"","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128436563","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906399
Huizhi Liu, Chen Li, Lihua Tian
In view of the problems of cold start and data interaction in recommendation systems, and most current recommendation algorithms ignore the diversity of data types, the combination of multimodal data and knowledge graph is bound to improve the pertinence of video recommendation. In this paper, we propose Multi-modal Knowledge Graph Attention Network (MMKGV) model, and all the entity nodes of the knowledge graph are innovatively introduced into multimodal information. The high-order recursive node information dissemination and information aggregation are carried out on the multimodal knowledge graph through the graph attention network. In the model, the triplet function of the knowledge graph is used to construct the triplet inference relationship, and the vector representation generated by the final aggregation is used for recommendation. Through extensive experiments on two public datasets TikTok and Kwai, the results show that the MMKGV can effectively improve the effect of video recommendation compared with other comparison algorithms.
{"title":"Multi-modal Graph Attention Network for Video Recommendation","authors":"Huizhi Liu, Chen Li, Lihua Tian","doi":"10.1109/CCET55412.2022.9906399","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906399","url":null,"abstract":"In view of the problems of cold start and data interaction in recommendation systems, and most current recommendation algorithms ignore the diversity of data types, the combination of multimodal data and knowledge graph is bound to improve the pertinence of video recommendation. In this paper, we propose Multi-modal Knowledge Graph Attention Network (MMKGV) model, and all the entity nodes of the knowledge graph are innovatively introduced into multimodal information. The high-order recursive node information dissemination and information aggregation are carried out on the multimodal knowledge graph through the graph attention network. In the model, the triplet function of the knowledge graph is used to construct the triplet inference relationship, and the vector representation generated by the final aggregation is used for recommendation. Through extensive experiments on two public datasets TikTok and Kwai, the results show that the MMKGV can effectively improve the effect of video recommendation compared with other comparison algorithms.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130605369","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906367
Min Wu, Chen Li, Lihua Tian
Thanks to the attention mechanism, self-attention networks (SANs) have been widely used in sequential recommendation. However, most existing SANs approaches still follow an old fashion generating one single embedding as final representation, which constraints model’s capacity. To enrich this kind of representation, sequential recommender uses metadata such as item category to capture user’s multi-interests. But this method will not reach its expectation due to item’s long-tail property. This property will result a large constant of category cannot be effectively activated by the lack of interaction records. Another drawback is that may also lead to over-parameterization caused by the massive categories. Particularly, we propose a Self-Attentive Interest Retrieval network (SAIR) to explore a context-aware representation from user’s behaviors while not fall into over-parameterization. SAIR works in a typical SANs manner, encode the behavior sequence using self-attention, and we propose an interest retrieval module to project the sequences to an interest relevance distribution adaptively. And we leverage an interest-to-interest interaction to generate several context-aware interests embeddings. Then we fuse multi-interest embeddings as final output. Extensive experiments are carried out on three real-world datasets, the results demonstrate that SAIR outperforms other SANs methods and other state-of-the-art algorithms in multiple evaluation metrics.
{"title":"A Self-Attentive Interest Retrieval Recommender","authors":"Min Wu, Chen Li, Lihua Tian","doi":"10.1109/CCET55412.2022.9906367","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906367","url":null,"abstract":"Thanks to the attention mechanism, self-attention networks (SANs) have been widely used in sequential recommendation. However, most existing SANs approaches still follow an old fashion generating one single embedding as final representation, which constraints model’s capacity. To enrich this kind of representation, sequential recommender uses metadata such as item category to capture user’s multi-interests. But this method will not reach its expectation due to item’s long-tail property. This property will result a large constant of category cannot be effectively activated by the lack of interaction records. Another drawback is that may also lead to over-parameterization caused by the massive categories. Particularly, we propose a Self-Attentive Interest Retrieval network (SAIR) to explore a context-aware representation from user’s behaviors while not fall into over-parameterization. SAIR works in a typical SANs manner, encode the behavior sequence using self-attention, and we propose an interest retrieval module to project the sequences to an interest relevance distribution adaptively. And we leverage an interest-to-interest interaction to generate several context-aware interests embeddings. Then we fuse multi-interest embeddings as final output. Extensive experiments are carried out on three real-world datasets, the results demonstrate that SAIR outperforms other SANs methods and other state-of-the-art algorithms in multiple evaluation metrics.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121544177","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906339
Feng Zhang, Yingqi Han, Jiong Wang, Jie Liu
Textual structure analysis is an important part of Automatic Essay Score (AES), and is also one of the important research directions in Natural Language Processing. At present, there are still deficiencies in the research of narrative textual structure in China, one of the main reasons is the lack of data available for research. To solve this problem, this paper proposes and constructs a corpus for the textual component identification of narrative essay. This paper divides the text structure of narrative essay, and forms a corpus for the narrative essay component identification. The paper finally annotated 3024 articles with 21128 sentences in total. This paper combines manual annotation and the automatic annotation of the model to build corpus, and conducts statistical analysis on the distribution of the corpus content and the consistency of the corpus annotation. The experiment shows text component recognition performance achieves 80.75% F 1 score. The work provided basic data for the research of AES.
{"title":"Construction of Narrative Text Component Recognition Corpus","authors":"Feng Zhang, Yingqi Han, Jiong Wang, Jie Liu","doi":"10.1109/CCET55412.2022.9906339","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906339","url":null,"abstract":"Textual structure analysis is an important part of Automatic Essay Score (AES), and is also one of the important research directions in Natural Language Processing. At present, there are still deficiencies in the research of narrative textual structure in China, one of the main reasons is the lack of data available for research. To solve this problem, this paper proposes and constructs a corpus for the textual component identification of narrative essay. This paper divides the text structure of narrative essay, and forms a corpus for the narrative essay component identification. The paper finally annotated 3024 articles with 21128 sentences in total. This paper combines manual annotation and the automatic annotation of the model to build corpus, and conducts statistical analysis on the distribution of the corpus content and the consistency of the corpus annotation. The experiment shows text component recognition performance achieves 80.75% F 1 score. The work provided basic data for the research of AES.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122680738","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906398
Qiwei Huan, B. Bai, Wenbiao Cai, Yundi Chen
In recent years, severe financing constraints restrict the development of SMEs (small and medium-sized enterprises) and the use of digital finance can play a greater role in the financing of SMEs. This paper conducts research through the empirical analysis method. Based on the existing research findings, the financing constraints of SMEs show a strong cash-cash flow sensitivity. So this paper build a cash-flow sensitivity model to measure financing constraints and add digital finance data variables to explore the impact of digital finance on SMEs’ financing constraints. Then Stata 15.0 statistical software was used to carry out descriptive statistics, correlation analysis, regression analysis, and robustness test. The results show that there are financing constraints in SMEs, and the financing constraints of SMEs are alleviated after joining the digital finance system, which shows that the financing constraints of SMEs can be effectively improved by the digital finance system, but the ability to innovate and empower needs to be improved. The development of digital finance can cut through the data islands, reduce moral risk, rich financing channels, and financing costs indirectly reduced.
{"title":"The Innovation of Financial Digitalization Enhances the Financing Capability of Small and Medium-Sized Enterprises: An Empirical Study","authors":"Qiwei Huan, B. Bai, Wenbiao Cai, Yundi Chen","doi":"10.1109/CCET55412.2022.9906398","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906398","url":null,"abstract":"In recent years, severe financing constraints restrict the development of SMEs (small and medium-sized enterprises) and the use of digital finance can play a greater role in the financing of SMEs. This paper conducts research through the empirical analysis method. Based on the existing research findings, the financing constraints of SMEs show a strong cash-cash flow sensitivity. So this paper build a cash-flow sensitivity model to measure financing constraints and add digital finance data variables to explore the impact of digital finance on SMEs’ financing constraints. Then Stata 15.0 statistical software was used to carry out descriptive statistics, correlation analysis, regression analysis, and robustness test. The results show that there are financing constraints in SMEs, and the financing constraints of SMEs are alleviated after joining the digital finance system, which shows that the financing constraints of SMEs can be effectively improved by the digital finance system, but the ability to innovate and empower needs to be improved. The development of digital finance can cut through the data islands, reduce moral risk, rich financing channels, and financing costs indirectly reduced.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123231662","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906364
Fan Xu, Bin Yin, Ming-Zhu Zhang, Xue Wang
As the aerospace business growing rapidly, data flow and volume has exploded in recent years, bringing chances and challenges to big data infrastructures and applications in this field. In traditional aerospace data and application centers, data is stored in network attached storages(NAS) and processed by sequential or low level parallel programs, which can hardly meet the demand of performance, availability and scalability. In this paper, we provided a big data infrastructure based on HDFS for big data centers, which can improve the availability and scalability remarkably. Besides, we gather a series of typical big data applications in aerospace filed as benchmarks, analyzes their characteristics and accelerates them in MapReduce framework. The experiment result shows that among all the benchmarks, the speedup is 4.98 to the peak and 3.87 on the average.
{"title":"Improving Aerospace Big Data Infrastructure and Applications with Distributed File System and Massive Parallel Calculation","authors":"Fan Xu, Bin Yin, Ming-Zhu Zhang, Xue Wang","doi":"10.1109/CCET55412.2022.9906364","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906364","url":null,"abstract":"As the aerospace business growing rapidly, data flow and volume has exploded in recent years, bringing chances and challenges to big data infrastructures and applications in this field. In traditional aerospace data and application centers, data is stored in network attached storages(NAS) and processed by sequential or low level parallel programs, which can hardly meet the demand of performance, availability and scalability. In this paper, we provided a big data infrastructure based on HDFS for big data centers, which can improve the availability and scalability remarkably. Besides, we gather a series of typical big data applications in aerospace filed as benchmarks, analyzes their characteristics and accelerates them in MapReduce framework. The experiment result shows that among all the benchmarks, the speedup is 4.98 to the peak and 3.87 on the average.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122673515","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 : 2022-08-19DOI: 10.1109/CCET55412.2022.9906382
Luchen Zhou, Tao Lu, X. Hu
As the important component of the Internet, the Web makes it easy for us to access information anytime, anywhere. However, the widespread adoption of web applications has introduced new security risks and expanded existing attack surfaces that many organizations are not effectively protecting. Among the various threats facing the web applications, injection attacks are one of the most dangerous. In this work, we propose to use one-class Support Vector Machine (SVM) for detecting web application injection attacks. We treat the detection of injection attacks as an anomaly detection problem. In the training stage, a number of legitimate HTTP requests are used to train a one-class SVM model. In the testing stage, the trained one-class SVM is used to detect whether an HTTP request is legitimate or malicious. We adopt 2v-gram algorithm (a variant of n-gram) to extract features from HTTP requests. The experimental results show that one-class SVM achieves good performance in detecting web application injection attacks by achieving a detection rate of 94.04% and a false positive rate of 1.62%.
{"title":"Detecting Web Application Injection Attacks Using One-Class SVM","authors":"Luchen Zhou, Tao Lu, X. Hu","doi":"10.1109/CCET55412.2022.9906382","DOIUrl":"https://doi.org/10.1109/CCET55412.2022.9906382","url":null,"abstract":"As the important component of the Internet, the Web makes it easy for us to access information anytime, anywhere. However, the widespread adoption of web applications has introduced new security risks and expanded existing attack surfaces that many organizations are not effectively protecting. Among the various threats facing the web applications, injection attacks are one of the most dangerous. In this work, we propose to use one-class Support Vector Machine (SVM) for detecting web application injection attacks. We treat the detection of injection attacks as an anomaly detection problem. In the training stage, a number of legitimate HTTP requests are used to train a one-class SVM model. In the testing stage, the trained one-class SVM is used to detect whether an HTTP request is legitimate or malicious. We adopt 2v-gram algorithm (a variant of n-gram) to extract features from HTTP requests. The experimental results show that one-class SVM achieves good performance in detecting web application injection attacks by achieving a detection rate of 94.04% and a false positive rate of 1.62%.","PeriodicalId":329327,"journal":{"name":"2022 IEEE 5th International Conference on Computer and Communication Engineering Technology (CCET)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132517683","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：