Pub Date : 2020-04-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202004007
张茁涵, 曹容玮, 李晨, 程士卿
To solve the problem of multi-view clustering,a latent low-rank sparse multi-view subspace clustering(LLSMSC)algorithm is proposed.A latent space shared by all views is constructed to explore the complementary information of multi-view data.The global and local structure of multi-view data can be captured to attain promising clustering results by imposing low-rank constraint and sparse constraint on the implicit latent subspace representation simultaneously.An algorithm based on augmented Lagrangian multiplier with alternating direction minimization strategy is employed to solve the optimization problem.Experiments on six benchmark datasets verify the effectiveness and superiority of LLSMSC.
{"title":"Latent Low-Rank Sparse Multi-view Subspace Clustering","authors":"张茁涵, 曹容玮, 李晨, 程士卿","doi":"10.16451/J.CNKI.ISSN1003-6059.202004007","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202004007","url":null,"abstract":"To solve the problem of multi-view clustering,a latent low-rank sparse multi-view subspace clustering(LLSMSC)algorithm is proposed.A latent space shared by all views is constructed to explore the complementary information of multi-view data.The global and local structure of multi-view data can be captured to attain promising clustering results by imposing low-rank constraint and sparse constraint on the implicit latent subspace representation simultaneously.An algorithm based on augmented Lagrangian multiplier with alternating direction minimization strategy is employed to solve the optimization problem.Experiments on six benchmark datasets verify the effectiveness and superiority of LLSMSC.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"344-352"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46486273","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 : 2020-04-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202004004
张荣国, 姚晓玲, 赵建, 胡静, 刘小君
To improve the accuracy and timeliness of spectral clustering image segmentation,an algorithm of manifold spectral clustering image segmentation based on local geometry features is proposed.Firstly,considering the manifold structure of image data,the relationship of data intrinsic dimensions is obtained by performing spectral clustering based on local principal components analysis in the k-nearest neighbor region of data points.Then,the local linear reconstruction technique in manifold learning is introduced,and the similarity of local tangent space between data is obtained via mixed linear analyzers,and the similarity matrix with local geometric features is constructed by merging the intrinsic dimension and the local tangent space.Nystr m technique is utilized to approximate eigenvectors of the image to be segmented,and spectral clustering is performed on the constructed k principal eigenvectors.Finally,experiments on Berkeley dataset show the advantages of the proposed algorithm in accuracy and timeliness.
{"title":"Manifold Spectral Clustering Image Segmentation Algorithm Based on Local Geometry Features","authors":"张荣国, 姚晓玲, 赵建, 胡静, 刘小君","doi":"10.16451/J.CNKI.ISSN1003-6059.202004004","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202004004","url":null,"abstract":"To improve the accuracy and timeliness of spectral clustering image segmentation,an algorithm of manifold spectral clustering image segmentation based on local geometry features is proposed.Firstly,considering the manifold structure of image data,the relationship of data intrinsic dimensions is obtained by performing spectral clustering based on local principal components analysis in the k-nearest neighbor region of data points.Then,the local linear reconstruction technique in manifold learning is introduced,and the similarity of local tangent space between data is obtained via mixed linear analyzers,and the similarity matrix with local geometric features is constructed by merging the intrinsic dimension and the local tangent space.Nystr m technique is utilized to approximate eigenvectors of the image to be segmented,and spectral clustering is performed on the constructed k principal eigenvectors.Finally,experiments on Berkeley dataset show the advantages of the proposed algorithm in accuracy and timeliness.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"313-324"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47264719","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 : 2020-04-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202004003
王军浩, 闫德勤, 刘德山, 闫汇聪
In joint sparse representation of hyperspectral image classification,once the local window of each pixel includes samples from different categories,the dictionary atoms and testing samples are easily affected by samples from different categories with same spectrum and the classification performance is seriously decreased.According to the characteristics of hyperspectral image,an algorithm of joint sparse representation fusing hierarchical deep network is proposed.Discriminative spectral information and spatial information are extracted by alternating spectral and spatial feature learning operations,and then a dictionary with spatial spectral features is constructed for joint sparse representation.In the classification process,the correlation coefficient between the dictionary and the testing samples is combined with classification error to make decisions.Experiments on two hyperspectral remote sensing datasets verify the effectiveness of the proposed algorithm.
{"title":"Joint Sparse Representation Fusing Hierarchical Deep Network of Hyperspectral Image Classification","authors":"王军浩, 闫德勤, 刘德山, 闫汇聪","doi":"10.16451/J.CNKI.ISSN1003-6059.202004003","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202004003","url":null,"abstract":"In joint sparse representation of hyperspectral image classification,once the local window of each pixel includes samples from different categories,the dictionary atoms and testing samples are easily affected by samples from different categories with same spectrum and the classification performance is seriously decreased.According to the characteristics of hyperspectral image,an algorithm of joint sparse representation fusing hierarchical deep network is proposed.Discriminative spectral information and spatial information are extracted by alternating spectral and spatial feature learning operations,and then a dictionary with spatial spectral features is constructed for joint sparse representation.In the classification process,the correlation coefficient between the dictionary and the testing samples is combined with classification error to make decisions.Experiments on two hyperspectral remote sensing datasets verify the effectiveness of the proposed algorithm.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"303-312"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42727711","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 : 2020-02-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202002008
刘昊鑫, 吴小俊, 庾骏
To solve the problem of low retrieval accuracy and long training time in cross-modal retrieval algorithms,a cross-modal retrieval algorithm joining hashing feature and classifier learning(HFCL)is proposed.Uniform hash codes are utilized to describe different modal data with the same semantics.In the training stage,label information is utilized to study discriminative hash codes.And the kernel logistic regression is adopted to learn the hash function of each modal.In the testing stage,for any sample,the hash feature is generated by learned hash function,and another modal datum related to its semantics is retrieved from the database.Experiments on three public datasets verify the effectiveness of HFCL.
{"title":"Joint Hashing Feature and Classifier Learning for Cross-Modal Retrieval","authors":"刘昊鑫, 吴小俊, 庾骏","doi":"10.16451/J.CNKI.ISSN1003-6059.202002008","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202002008","url":null,"abstract":"To solve the problem of low retrieval accuracy and long training time in cross-modal retrieval algorithms,a cross-modal retrieval algorithm joining hashing feature and classifier learning(HFCL)is proposed.Uniform hash codes are utilized to describe different modal data with the same semantics.In the training stage,label information is utilized to study discriminative hash codes.And the kernel logistic regression is adopted to learn the hash function of each modal.In the testing stage,for any sample,the hash feature is generated by learned hash function,and another modal datum related to its semantics is retrieved from the database.Experiments on three public datasets verify the effectiveness of HFCL.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"160-165"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48565786","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 : 2020-02-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202002001
彭鹏, 倪志伟, 朱旭辉, 夏平凡
Aiming at the problems of dimension reduction and redundancy removing,an attribute reduction method based on improved binary glowworm swarm optimization algorithm and neighborhood rough set is proposed.Firstly,the population is collaborative initialization using reverse learning,and the mapping of the change function based on Sigmoid is employed for binary coding,and an improved binary glowworm opti-mization algorithm is proposed with Levy flight position update strategy.Secondly,neighborhood rough set is employed as an evaluation criterion,and the proposed algorithm is utilized as an search strategy for attribute reduction.Finally,experiments on the standard UCI datasets demonstrate the effectiveness of the attribute reduction method,and the better convergence speed and accuracy of the proposed algorithm is verified.
{"title":"Attribute Reduction Method Based on Improved Binary Glowworm Swarm Optimization Algorithm and Neighborhood Rough Set","authors":"彭鹏, 倪志伟, 朱旭辉, 夏平凡","doi":"10.16451/J.CNKI.ISSN1003-6059.202002001","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202002001","url":null,"abstract":"Aiming at the problems of dimension reduction and redundancy removing,an attribute reduction method based on improved binary glowworm swarm optimization algorithm and neighborhood rough set is proposed.Firstly,the population is collaborative initialization using reverse learning,and the mapping of the change function based on Sigmoid is employed for binary coding,and an improved binary glowworm opti-mization algorithm is proposed with Levy flight position update strategy.Secondly,neighborhood rough set is employed as an evaluation criterion,and the proposed algorithm is utilized as an search strategy for attribute reduction.Finally,experiments on the standard UCI datasets demonstrate the effectiveness of the attribute reduction method,and the better convergence speed and accuracy of the proposed algorithm is verified.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"95-105"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48608869","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 : 2020-02-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202002007
何丽, 韩克平, 朱泓西, 刘颖
To solve the catastrophic forgetting problem caused by incremental learning,a deep incremental image classification method based on double-branch iteration is proposed.The primary network is utilized to store the acquired old class knowledge,while the branch network is exploited to learn the new class knowledge.The parameters of the branch network are optimized by the weight of the primary network in the incremental iteration process.Density peak clustering method is employed to select typical samples from the iterative dataset and construct retention set.The retention set is added into the incremental iteration training to mitigate catastrophic forgetting.The experiments demonstrate the better performance of the proposed method.
{"title":"Deep Incremental Image Classification Method Based on Double-Branch Iteration","authors":"何丽, 韩克平, 朱泓西, 刘颖","doi":"10.16451/J.CNKI.ISSN1003-6059.202002007","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202002007","url":null,"abstract":"To solve the catastrophic forgetting problem caused by incremental learning,a deep incremental image classification method based on double-branch iteration is proposed.The primary network is utilized to store the acquired old class knowledge,while the branch network is exploited to learn the new class knowledge.The parameters of the branch network are optimized by the weight of the primary network in the incremental iteration process.Density peak clustering method is employed to select typical samples from the iterative dataset and construct retention set.The retention set is added into the incremental iteration training to mitigate catastrophic forgetting.The experiments demonstrate the better performance of the proposed method.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"150-159"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41687011","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 : 2020-02-01DOI: 10.16451/J.CNKI.ISSN1003-6059.202002002
苏立新, 郭嘉丰, 范意兴, 兰艳艳, 程学旗
In the existing extractive reading comprehension models,only the boundary of answers is utilized as the supervision signal and the labeling processed by human is ignored.Consequently,learned models are prone to learn the superficial features and the generalization performance is degraded.In this paper,a label-enhanced reading comprehension model is proposed to imitate human activity.The answer-bearing sentence,the content and the boundary of the answer are learned simultaneously.The answer-bearing sentence and the content of the answer can be derived from the boundary of the answer and these three types of labels are regarded as supervision signals.The model is trained by multitask learning.During prediction,the probabilities from three predictions are merged to determine the answer,and thus the generalization performance is improved.Experiments on SQuAD dataset demonstrate the effectiveness of LE-Reader model.
{"title":"Label-Enhanced Reading Comprehension Model","authors":"苏立新, 郭嘉丰, 范意兴, 兰艳艳, 程学旗","doi":"10.16451/J.CNKI.ISSN1003-6059.202002002","DOIUrl":"https://doi.org/10.16451/J.CNKI.ISSN1003-6059.202002002","url":null,"abstract":"In the existing extractive reading comprehension models,only the boundary of answers is utilized as the supervision signal and the labeling processed by human is ignored.Consequently,learned models are prone to learn the superficial features and the generalization performance is degraded.In this paper,a label-enhanced reading comprehension model is proposed to imitate human activity.The answer-bearing sentence,the content and the boundary of the answer are learned simultaneously.The answer-bearing sentence and the content of the answer can be derived from the boundary of the answer and these three types of labels are regarded as supervision signals.The model is trained by multitask learning.During prediction,the probabilities from three predictions are merged to determine the answer,and thus the generalization performance is improved.Experiments on SQuAD dataset demonstrate the effectiveness of LE-Reader model.","PeriodicalId":34917,"journal":{"name":"模式识别与人工智能","volume":"33 1","pages":"106-112"},"PeriodicalIF":0.0,"publicationDate":"2020-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46066186","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：