Pub Date : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00063
C. Fan, Yongchun Li, Yuexin Wu
In recent years, the number of published scientific research papers has shown a growing trend. How to classify scientific research papers efficiently and accurately is a very important issue. However, excellent paper classification system platforms at home and abroad, such as China National Knowledge Infrastructure, Microsoft Academic Network, etc., rely heavily on the structured or semi-structured text in papers for classification, and do not interpret the unstructured text data in papers enough. To solve this problem, we proposed a multi-feature fusion paper classification model based on attention mechanism (AttentionMFF), which uses the fusion features of structured and unstructured text data in papers to improve classification performance. First, Attention MFF extracts the features of different texts in papers by a BERT layer, then uses attention mechanism to fuse different features, and finally get category through the linear layer. Experiments on the arXiv paper dataset show that the Attention MFF has higher F1-Score than TextCNN model and BERT model that only uses the feature of abstract.
{"title":"Multi feature fusion paper classification model based on attention mechanism","authors":"C. Fan, Yongchun Li, Yuexin Wu","doi":"10.1109/ICNLP58431.2023.00063","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00063","url":null,"abstract":"In recent years, the number of published scientific research papers has shown a growing trend. How to classify scientific research papers efficiently and accurately is a very important issue. However, excellent paper classification system platforms at home and abroad, such as China National Knowledge Infrastructure, Microsoft Academic Network, etc., rely heavily on the structured or semi-structured text in papers for classification, and do not interpret the unstructured text data in papers enough. To solve this problem, we proposed a multi-feature fusion paper classification model based on attention mechanism (AttentionMFF), which uses the fusion features of structured and unstructured text data in papers to improve classification performance. First, Attention MFF extracts the features of different texts in papers by a BERT layer, then uses attention mechanism to fuse different features, and finally get category through the linear layer. Experiments on the arXiv paper dataset show that the Attention MFF has higher F1-Score than TextCNN model and BERT model that only uses the feature of abstract.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75240793","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}
Given incomplete source sentences as inputs, it is generally difficult for Simultaneous Machine Translation (SiMT) models to generate a target token once its aligned source tokens are absent. How to measure such difficulty and further conduct adaptive training for SiMT models are not sufficiently studied. In this paper, we propose a new metric named alignment offset (AO) to quantify the learning difficulty of target tokens for SiMT models. Given a target token, its AO is calculated by the offset between its aligned source tokens and the already received source tokens. Furthermore, we design two AO-based adaptive training methods to improve the training of SiMT models. Firstly, we introduce token-level curriculum learning based on AO, which progressively switches the training process from easy target tokens to difficult ones. Secondly, we assign an appropriate weight to the training loss of each target token according to its AO. Experimental results on four datasets demonstrate that our methods significantly and consistently outperform all the strong baselines.
{"title":"Alignment Offset Based Adaptive Training for Simultaneous Machine Translation","authors":"Qiqi Liang, Yanjun Liu, Fandong Meng, Jinan Xu, Yufeng Chen, Jie Zhou","doi":"10.1109/ICNLP58431.2023.00035","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00035","url":null,"abstract":"Given incomplete source sentences as inputs, it is generally difficult for Simultaneous Machine Translation (SiMT) models to generate a target token once its aligned source tokens are absent. How to measure such difficulty and further conduct adaptive training for SiMT models are not sufficiently studied. In this paper, we propose a new metric named alignment offset (AO) to quantify the learning difficulty of target tokens for SiMT models. Given a target token, its AO is calculated by the offset between its aligned source tokens and the already received source tokens. Furthermore, we design two AO-based adaptive training methods to improve the training of SiMT models. Firstly, we introduce token-level curriculum learning based on AO, which progressively switches the training process from easy target tokens to difficult ones. Secondly, we assign an appropriate weight to the training loss of each target token according to its AO. Experimental results on four datasets demonstrate that our methods significantly and consistently outperform all the strong baselines.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85477397","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00016
Lan Rong, Haowen Mi, Qu Na, Zhao Feng, Haiyan Yu, Zhang Lu
Although the evidence c-means clustering (ECM) has the capability to process uncertain information, it is not suitable for noisy image segmentation, because the spatial information of pixels is not considered. To solve the problem, an adaptive kernelized evidence c-means clustering combining spatial information for noisy image segmentation algorithm is proposed. Firstly, an adaptive noise distance that can be iteratively updated is constructed using the local information of the pixels. Secondly, to improve the classification performance, an adaptive kernel function is proposed to measure the distance between the pixel and the cluster center. Simultaneously, the original, local and non-local information of pixels are introduced adaptively into the objective function to enhance the robustness to noise. In the iteration, the noise cluster is automatically recovered using the recovery factor constructed by the gray and spatial information of neighborhood pixels. Finally, the credal partition is transformed into a fuzzy partition by pignistic transformation, the classification of pixel be determined by the maximum membership principle. Experiments on synthetic images and real images demonstrate that the proposed algorithm has strong noise suppression ability. Visual effects and evaluation indexes verify the effectiveness of the proposed algorithm for noisy image segmentation.
{"title":"Adaptive Kernelized Evidence C-Means Clustering Combining Spatial Information for Noisy Image Segmentation","authors":"Lan Rong, Haowen Mi, Qu Na, Zhao Feng, Haiyan Yu, Zhang Lu","doi":"10.1109/ICNLP58431.2023.00016","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00016","url":null,"abstract":"Although the evidence c-means clustering (ECM) has the capability to process uncertain information, it is not suitable for noisy image segmentation, because the spatial information of pixels is not considered. To solve the problem, an adaptive kernelized evidence c-means clustering combining spatial information for noisy image segmentation algorithm is proposed. Firstly, an adaptive noise distance that can be iteratively updated is constructed using the local information of the pixels. Secondly, to improve the classification performance, an adaptive kernel function is proposed to measure the distance between the pixel and the cluster center. Simultaneously, the original, local and non-local information of pixels are introduced adaptively into the objective function to enhance the robustness to noise. In the iteration, the noise cluster is automatically recovered using the recovery factor constructed by the gray and spatial information of neighborhood pixels. Finally, the credal partition is transformed into a fuzzy partition by pignistic transformation, the classification of pixel be determined by the maximum membership principle. Experiments on synthetic images and real images demonstrate that the proposed algorithm has strong noise suppression ability. Visual effects and evaluation indexes verify the effectiveness of the proposed algorithm for noisy image segmentation.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76723762","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 : 2023-03-01DOI: 10.1109/icnlp58431.2023.00034
Youyao Liu, Jiale Chen, Jialei Gao, Shihao Gai
Artificial intelligence is the vane leading the world’s scientific and technological development and future lifestyle change in the 21st century, and speech recognition, as one of the indispensable technical means, is inevitably the focus of human attention. There are two problems in traditional speech recognition: first, speech recognition technology cannot be significantly improved, and second, speech recognition systems cannot accurately extract data and features. In order to solve these problems, this paper first compares the traditional speech recognition GMM-HMM model and establishes a DNN-HMM model, which proposes a method to improve the speed of speech recognition and greatly improves the recognition rate. However, DNN-HMM lacks the ability to use historical information to assist in the current task, and a second model is proposed on the basis of this problem, that is, the LSTM model is used to solve the problem of insufficient contextual information, which further improves the speech recognition ability. Then, in order to solve the problem of long memory loss and speed up training, the Transformer model is cited, and in order to solve the problem that the traditional language model can only predict the next word in one direction, the BERT model, which has a bidirectional language model, is invoked.
{"title":"A Survey of Speech Recognition Based on Deep Learning","authors":"Youyao Liu, Jiale Chen, Jialei Gao, Shihao Gai","doi":"10.1109/icnlp58431.2023.00034","DOIUrl":"https://doi.org/10.1109/icnlp58431.2023.00034","url":null,"abstract":"Artificial intelligence is the vane leading the world’s scientific and technological development and future lifestyle change in the 21st century, and speech recognition, as one of the indispensable technical means, is inevitably the focus of human attention. There are two problems in traditional speech recognition: first, speech recognition technology cannot be significantly improved, and second, speech recognition systems cannot accurately extract data and features. In order to solve these problems, this paper first compares the traditional speech recognition GMM-HMM model and establishes a DNN-HMM model, which proposes a method to improve the speed of speech recognition and greatly improves the recognition rate. However, DNN-HMM lacks the ability to use historical information to assist in the current task, and a second model is proposed on the basis of this problem, that is, the LSTM model is used to solve the problem of insufficient contextual information, which further improves the speech recognition ability. Then, in order to solve the problem of long memory loss and speed up training, the Transformer model is cited, and in order to solve the problem that the traditional language model can only predict the next word in one direction, the BERT model, which has a bidirectional language model, is invoked.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77048846","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00014
Sugang Ma, Siwei Sun, Lei Pu, Xiaobao Yang
To address the problem of insufficient representational capability and lack of online update of the Fully-convolutional Siamese Network (SiamFC) tracker in complex scenes, this paper proposes a siamese network visual tracking algorithm based on GCT attention and dual-template update mechanism. First, the feature extraction network is constructed by replacing AlexNet with the VGG16 network and SoftPool is used to replace the maximum pooling layer. Secondly, the attention module is added after the backbone network to enhance the network’s ability to extract object features. Finally, a dual-template update mechanism is designed for response map fusion. Average Peak-to-Correlation Energy (APCE) is used to determine whether to update the dynamic templates, effectively improving the tracking robustness. The proposed algorithm is trained on the Got-10k dataset and tested on the OTB2015 and VOT2018 datasets. The experimental results show that, compared with SiamFC, the success rate and accuracy reach 0.663 and 0.891 on the OTB2015, which improve respectively 7.6% and 11.9%; On the VOT2018 dataset, the tracking accuracy, robustness and EAO are improved respectively by 2.9%, 29% and 14%. The proposed algorithm achieves high tracking accuracy in complex scenes and the tracking speed reaches 52.6 Fps, which meets the real-time tracking requirements.
{"title":"Siamese Network Visual Tracking Algorithm Based on GCT Attention and Dual-Template Update","authors":"Sugang Ma, Siwei Sun, Lei Pu, Xiaobao Yang","doi":"10.1109/ICNLP58431.2023.00014","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00014","url":null,"abstract":"To address the problem of insufficient representational capability and lack of online update of the Fully-convolutional Siamese Network (SiamFC) tracker in complex scenes, this paper proposes a siamese network visual tracking algorithm based on GCT attention and dual-template update mechanism. First, the feature extraction network is constructed by replacing AlexNet with the VGG16 network and SoftPool is used to replace the maximum pooling layer. Secondly, the attention module is added after the backbone network to enhance the network’s ability to extract object features. Finally, a dual-template update mechanism is designed for response map fusion. Average Peak-to-Correlation Energy (APCE) is used to determine whether to update the dynamic templates, effectively improving the tracking robustness. The proposed algorithm is trained on the Got-10k dataset and tested on the OTB2015 and VOT2018 datasets. The experimental results show that, compared with SiamFC, the success rate and accuracy reach 0.663 and 0.891 on the OTB2015, which improve respectively 7.6% and 11.9%; On the VOT2018 dataset, the tracking accuracy, robustness and EAO are improved respectively by 2.9%, 29% and 14%. The proposed algorithm achieves high tracking accuracy in complex scenes and the tracking speed reaches 52.6 Fps, which meets the real-time tracking requirements.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77884382","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}
Retrieval-based dialogue systems have achieved great performance improvements after the raise of pre-trained language models and Transformer mechanisms. In the process of context and response selection, the pre-trained language model can capture the relationship between texts, but current existing methods don’t consider the order of sentences and the relationship between the context and the response. At the same time, as the problem of a small number of positive samples in retrieval-based dialogue systems, it is difficult to train a learning model with high performance. In addition, existing methods usually requires the larger computational cost after splicing the context and the response. To solve the above problems, we propose a post-encoding approach combining with the strategy of contrastive learning. The order of the context and the relationship between sentences in dialogues and response are reflected in the encoding process, and a new loss function is designed for contrastive learning. The propose approach is validated through experiments on public datasets. The experiment results show that our model achieves better performance and effectiveness compared to existing methods.
{"title":"Post-encoding and contrastive learning method for response selection task","authors":"Xianwei Xue, Chunping Li, Zhilin Lu, Youshu Zhang, Shanghua Xiao","doi":"10.1109/ICNLP58431.2023.00050","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00050","url":null,"abstract":"Retrieval-based dialogue systems have achieved great performance improvements after the raise of pre-trained language models and Transformer mechanisms. In the process of context and response selection, the pre-trained language model can capture the relationship between texts, but current existing methods don’t consider the order of sentences and the relationship between the context and the response. At the same time, as the problem of a small number of positive samples in retrieval-based dialogue systems, it is difficult to train a learning model with high performance. In addition, existing methods usually requires the larger computational cost after splicing the context and the response. To solve the above problems, we propose a post-encoding approach combining with the strategy of contrastive learning. The order of the context and the relationship between sentences in dialogues and response are reflected in the encoding process, and a new loss function is designed for contrastive learning. The propose approach is validated through experiments on public datasets. The experiment results show that our model achieves better performance and effectiveness compared to existing methods.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87773012","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 : 2023-03-01DOI: 10.1109/icnlp58431.2023.00017
Lan Rong, Danlin Feng, Zhao Feng, Haiyan Yu, Zhang Lu
In order to make full use of the color information of the image and improve the accuracy of color image segmentation, this paper proposes an adaptive SLIC-based fuzzy intensity dissimilarity thresholding for color image segmentation, which does not need gray conversion. Firstly, the proposed algorithm adaptively selects the number of super-pixels through the sum of image information and image complexity, and uses SLIC technology to extract image super-pixels; Then, the median value of each channel pixel in each super-pixel block is used as the super-pixel value to calculate the super-pixel intensity information, and the super-pixel intensity histogram is counted; Finally, an intensity dissimilarity function based on IT2FS is constructed to search the optimal threshold. On Berkeley images and Weizmann images, the proposed algorithm is compared with the five related algorithms. The experiments show that the proposed algorithm has achieved good results in terms of visual effects and evaluation indicators, which proves the effectiveness of the algorithm.
{"title":"Adaptive SLIC-Based Fuzzy Intensity Dissimilarity Thresholding for Color Image Segmentation","authors":"Lan Rong, Danlin Feng, Zhao Feng, Haiyan Yu, Zhang Lu","doi":"10.1109/icnlp58431.2023.00017","DOIUrl":"https://doi.org/10.1109/icnlp58431.2023.00017","url":null,"abstract":"In order to make full use of the color information of the image and improve the accuracy of color image segmentation, this paper proposes an adaptive SLIC-based fuzzy intensity dissimilarity thresholding for color image segmentation, which does not need gray conversion. Firstly, the proposed algorithm adaptively selects the number of super-pixels through the sum of image information and image complexity, and uses SLIC technology to extract image super-pixels; Then, the median value of each channel pixel in each super-pixel block is used as the super-pixel value to calculate the super-pixel intensity information, and the super-pixel intensity histogram is counted; Finally, an intensity dissimilarity function based on IT2FS is constructed to search the optimal threshold. On Berkeley images and Weizmann images, the proposed algorithm is compared with the five related algorithms. The experiments show that the proposed algorithm has achieved good results in terms of visual effects and evaluation indicators, which proves the effectiveness of the algorithm.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83919324","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00046
Katsunori Kotani, T. Yoshimi
Learner corpus research has revealed the appropriateness of language learners’ language use by analyzing corpus data including annotation tags. Annotation tags provide linguistic information such as part-of-speech and error information on lexical, syntactic, semantic, and phonetic items. Recent learner corpus research compiling a multimodal learner corpus has extended the research target to nonverbal behaviors such as facial expressions and gesturing because nonverbal behaviors play a significant role in communication. The goal of this paper is two-fold. The first objective is to validate nonverbal-behavior annotation tags of the previous multimodal learner corpora. The second objective is to propose a plausible nonverbal-behavior tag set for a multimodal learner corpus.
{"title":"Assessment of Nonverbal-behavior Annotation Tags in Multimodal Learner Corpus","authors":"Katsunori Kotani, T. Yoshimi","doi":"10.1109/ICNLP58431.2023.00046","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00046","url":null,"abstract":"Learner corpus research has revealed the appropriateness of language learners’ language use by analyzing corpus data including annotation tags. Annotation tags provide linguistic information such as part-of-speech and error information on lexical, syntactic, semantic, and phonetic items. Recent learner corpus research compiling a multimodal learner corpus has extended the research target to nonverbal behaviors such as facial expressions and gesturing because nonverbal behaviors play a significant role in communication. The goal of this paper is two-fold. The first objective is to validate nonverbal-behavior annotation tags of the previous multimodal learner corpora. The second objective is to propose a plausible nonverbal-behavior tag set for a multimodal learner corpus.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88692089","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 : 2023-03-01DOI: 10.1109/ICNLP58431.2023.00067
Kunpeng Zhang, Guangyue Lu, Yuxin Li, Cai Xu
Anomaly detection in attributed networks aims to find anomalous nodes in the network that differ from the behavior pattern of most nodes, and graph neural network provide a way to use fused structural and attribute information. However, existing methods based on Graph Convolutional Network (GCN) detection do not consider the over-smoothing phenomenon of GCN due to the stacks of network layers, which causes significant performance deterioration. To address the above problems, we propose a graph autoencoder-based anomaly detection method for attributed networks: Residual Graph Autoencoder (Res-GAE), by which the performance is effectively improved. Res-GAE contains an encoder and two decoders. More specifically, the encoder consists of a GCN and a residual network is utilized to learn the network representation. The decoders are designed to reconstruct the network structure and node attributes respectively. After that, the objective function is used to analyze the reconstruction error to generate the anomaly score ranking, to realize anomaly detection. Extensive experiments on the three datasets (BlogCatalog, Flickr, ACM) demonstrate that the proposed method has the significant improvement compared with other baseline methods.
{"title":"A Graph Autoencoder-based Anomaly Detection Method for Attributed Networks","authors":"Kunpeng Zhang, Guangyue Lu, Yuxin Li, Cai Xu","doi":"10.1109/ICNLP58431.2023.00067","DOIUrl":"https://doi.org/10.1109/ICNLP58431.2023.00067","url":null,"abstract":"Anomaly detection in attributed networks aims to find anomalous nodes in the network that differ from the behavior pattern of most nodes, and graph neural network provide a way to use fused structural and attribute information. However, existing methods based on Graph Convolutional Network (GCN) detection do not consider the over-smoothing phenomenon of GCN due to the stacks of network layers, which causes significant performance deterioration. To address the above problems, we propose a graph autoencoder-based anomaly detection method for attributed networks: Residual Graph Autoencoder (Res-GAE), by which the performance is effectively improved. Res-GAE contains an encoder and two decoders. More specifically, the encoder consists of a GCN and a residual network is utilized to learn the network representation. The decoders are designed to reconstruct the network structure and node attributes respectively. After that, the objective function is used to analyze the reconstruction error to generate the anomaly score ranking, to realize anomaly detection. Extensive experiments on the three datasets (BlogCatalog, Flickr, ACM) demonstrate that the proposed method has the significant improvement compared with other baseline methods.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89892466","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 : 2023-03-01DOI: 10.1109/icnlp58431.2023.00021
Tingting Chen, Xiaodong Zhang, Meixia Miao, Pengfei Tu
In order to solve the problems of simple chaotic behavior, poor initial value sensitivity corresponding to the traditional one-dimensional chaotic maps, a one-dimensional combined chaotic map based on fuzzy entropy theory is proposed. The designed combinatorial chaotic map combines the definition of fuzzy entropy and the classical one-dimensional chaotic map, and can effectively improve the performance of the combinatorial chaotic maps by extending the system parameters. Through the experimental simulation and analysis of the bifurcation diagram, Lyapunov exponent, initial value sensitivity and other related properties of the combined chaotic map, the experimental results show that the combined chaotic map constructed in this paper has good chaotic properties such as initial value sensitivity and randomness.
{"title":"Construction and Performance Analysis of Combinatorial Chaotic Map Based on Fuzzy Entropy","authors":"Tingting Chen, Xiaodong Zhang, Meixia Miao, Pengfei Tu","doi":"10.1109/icnlp58431.2023.00021","DOIUrl":"https://doi.org/10.1109/icnlp58431.2023.00021","url":null,"abstract":"In order to solve the problems of simple chaotic behavior, poor initial value sensitivity corresponding to the traditional one-dimensional chaotic maps, a one-dimensional combined chaotic map based on fuzzy entropy theory is proposed. The designed combinatorial chaotic map combines the definition of fuzzy entropy and the classical one-dimensional chaotic map, and can effectively improve the performance of the combinatorial chaotic maps by extending the system parameters. Through the experimental simulation and analysis of the bifurcation diagram, Lyapunov exponent, initial value sensitivity and other related properties of the combined chaotic map, the experimental results show that the combined chaotic map constructed in this paper has good chaotic properties such as initial value sensitivity and randomness.","PeriodicalId":53637,"journal":{"name":"Icon","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78193234","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：