Pub Date : 2018-10-01DOI: 10.1109/BRACIS.2018.00030
Cyntia Eico Hayama Nishida, Anna Helena Reali Costa, R. Bianchi
Basin of attraction contains biological functions and channels cell behavior, so when a gene network is in an unhealthy basin it may cause diseases. Control techniques can support the design of therapies that promote the transition of a biological system from diseased to healthier basins. Most control methods first infer a gene network and then derive a control strategy to avoid diseased states. However, this approach is limited to few genes and may cause other diseases, as the biological side of the problem is not considered. While changing between basins may change a diseased biological function for a healthier one, state avoidance can change functions in an unexpected way. We propose to extend a batch reinforcement learning method FQI-Sarsa, to change basin of attractions in a partial observable network. Using a batch reinforcement learning technique avoids the most time consuming phases that are the inference and control of the gene network. Results demonstrate that our method, BOAFQI-Sarsa, is more effective than previous studies that do not consider basins in their computations.
{"title":"Control of Gene Regulatory Networks Basin of Attractions with Batch Reinforcement Learning","authors":"Cyntia Eico Hayama Nishida, Anna Helena Reali Costa, R. Bianchi","doi":"10.1109/BRACIS.2018.00030","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00030","url":null,"abstract":"Basin of attraction contains biological functions and channels cell behavior, so when a gene network is in an unhealthy basin it may cause diseases. Control techniques can support the design of therapies that promote the transition of a biological system from diseased to healthier basins. Most control methods first infer a gene network and then derive a control strategy to avoid diseased states. However, this approach is limited to few genes and may cause other diseases, as the biological side of the problem is not considered. While changing between basins may change a diseased biological function for a healthier one, state avoidance can change functions in an unexpected way. We propose to extend a batch reinforcement learning method FQI-Sarsa, to change basin of attractions in a partial observable network. Using a batch reinforcement learning technique avoids the most time consuming phases that are the inference and control of the gene network. Results demonstrate that our method, BOAFQI-Sarsa, is more effective than previous studies that do not consider basins in their computations.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120972818","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00104
Xuejian Wang, M. Fairhurst, A. Canuto
Facial expressions can be seen as a form of non-verbal communication as well as a primary means of conveying social information among humans.Automatic facial expression recognition (FER) can be applied to a wide range of scenarios in human-computer interaction, facial animation, entertainment, and psychology studies. For feature representation in a FER system, various texture descriptors have been employed to derive an effective solution for this system. However, these individual texture descriptor-based FER systems have often failed to achieve effective performance in the recognition of facial expressions. In this sense, it is necessary to further improve the general performance of a facial expression recognition system, evaluating different feature representations. In this paper, a novel local descriptor for a facial expression recognition system is proposed, designated the level of difference descriptor (LOD). The main goal is to use this descriptor as a supplement to state-of-the-art local descriptors to further improve the performance of a FER system in terms of classification accuracy. Furthermore, the fusion of various texture features for devising a robust feature representation for multi-view facial expression recognition is presented.
{"title":"Fusion of Local Descriptors for Multi-view Facial Expression Recognition","authors":"Xuejian Wang, M. Fairhurst, A. Canuto","doi":"10.1109/BRACIS.2018.00104","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00104","url":null,"abstract":"Facial expressions can be seen as a form of non-verbal communication as well as a primary means of conveying social information among humans.Automatic facial expression recognition (FER) can be applied to a wide range of scenarios in human-computer interaction, facial animation, entertainment, and psychology studies. For feature representation in a FER system, various texture descriptors have been employed to derive an effective solution for this system. However, these individual texture descriptor-based FER systems have often failed to achieve effective performance in the recognition of facial expressions. In this sense, it is necessary to further improve the general performance of a facial expression recognition system, evaluating different feature representations. In this paper, a novel local descriptor for a facial expression recognition system is proposed, designated the level of difference descriptor (LOD). The main goal is to use this descriptor as a supplement to state-of-the-art local descriptors to further improve the performance of a FER system in terms of classification accuracy. Furthermore, the fusion of various texture features for devising a robust feature representation for multi-view facial expression recognition is presented.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116703874","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00100
V. A. Padilha, A. Carvalho
Biclustering algorithms have become one of the main tools for the analysis of gene expression data. They allow the identification of local patterns defined by subsets of genes and subsets of samples, which cannot be detected by traditional clustering algorithms. However, although useful, biclustering is a NP-hard problem. Therefore, the majority of biclustering algorithms look for biclusters optimizing a pre-established coherence measure. In the last 20 years, several heuristics and measures have been published for biclustering. However, most of these publications do not provide an extensive comparison of bicluster coherence measures on practical scenarios. To deal with this problem, this paper analyze the behavior of 15 bicluster coherence measures and external evaluation regarding 9 algorithms from the literature on gene expression datasets. According to the experimental results, there is no clear relation between these measures and assessment using information from gene ontology.
{"title":"A Study of Biclustering Coherence Measures for Gene Expression Data","authors":"V. A. Padilha, A. Carvalho","doi":"10.1109/BRACIS.2018.00100","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00100","url":null,"abstract":"Biclustering algorithms have become one of the main tools for the analysis of gene expression data. They allow the identification of local patterns defined by subsets of genes and subsets of samples, which cannot be detected by traditional clustering algorithms. However, although useful, biclustering is a NP-hard problem. Therefore, the majority of biclustering algorithms look for biclusters optimizing a pre-established coherence measure. In the last 20 years, several heuristics and measures have been published for biclustering. However, most of these publications do not provide an extensive comparison of bicluster coherence measures on practical scenarios. To deal with this problem, this paper analyze the behavior of 15 bicluster coherence measures and external evaluation regarding 9 algorithms from the literature on gene expression datasets. According to the experimental results, there is no clear relation between these measures and assessment using information from gene ontology.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125195472","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00062
Edesio Alcobaça, R. G. Mantovani, A. L. Rossi, A. Carvalho
Given the increase in data generation, as many algorithms have become available in recent years, the algorithm recommendation problem has attracted increasing attention in Machine Learning. This problem has been addressed in the Machine Learning community as a learning task at the meta-level where the most suitable algorithm has to be recommended for a specific dataset. Since it is not trivial to define which characteristics are the most useful for a specific domain, several meta-features have been proposed and used, increasing the meta-data meta-feature dimension. This study investigates the influence of dimensionality reduction techniques on the quality of the algorithm recommendation process. Experiments were carried out with 15 algorithm recommendation problems from the Aslib library, 4 meta-learners, and 3 dimensionality reduction techniques. The experimental results showed that linear aggregation techniques, such as PCA and LDA, can be used in algorithm recommendation problems to reduce the number of meta-features and computational cost without losing predictive performance.
{"title":"Dimensionality Reduction for the Algorithm Recommendation Problem","authors":"Edesio Alcobaça, R. G. Mantovani, A. L. Rossi, A. Carvalho","doi":"10.1109/BRACIS.2018.00062","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00062","url":null,"abstract":"Given the increase in data generation, as many algorithms have become available in recent years, the algorithm recommendation problem has attracted increasing attention in Machine Learning. This problem has been addressed in the Machine Learning community as a learning task at the meta-level where the most suitable algorithm has to be recommended for a specific dataset. Since it is not trivial to define which characteristics are the most useful for a specific domain, several meta-features have been proposed and used, increasing the meta-data meta-feature dimension. This study investigates the influence of dimensionality reduction techniques on the quality of the algorithm recommendation process. Experiments were carried out with 15 algorithm recommendation problems from the Aslib library, 4 meta-learners, and 3 dimensionality reduction techniques. The experimental results showed that linear aggregation techniques, such as PCA and LDA, can be used in algorithm recommendation problems to reduce the number of meta-features and computational cost without losing predictive performance.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121742678","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00033
J. E. H. D. Silva, H. Bernardino
The development of an efficient crossover for Cartesian Genetic Programming (CGP) has been widely investigated, but there is not a large number of approaches using this type of operator when designing combinational logic circuits. In this paper, we introduce a new crossover for CGP when using a single genotype representation and the desired model has multiple outputs. The proposal modifies the standard evolutionary strategy commonly adopted in CGP by combining the subgraphs of the best outputs of the parent and its offspring in order to generate a new fittest individual. The proposed crossover is applied to combinational logic circuits with multiple outputs, a parameter analysis is performed, and the results obtained are compared to those found by a baseline CGP and other techniques from the literature.
{"title":"Cartesian Genetic Programming with Crossover for Designing Combinational Logic Circuits","authors":"J. E. H. D. Silva, H. Bernardino","doi":"10.1109/BRACIS.2018.00033","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00033","url":null,"abstract":"The development of an efficient crossover for Cartesian Genetic Programming (CGP) has been widely investigated, but there is not a large number of approaches using this type of operator when designing combinational logic circuits. In this paper, we introduce a new crossover for CGP when using a single genotype representation and the desired model has multiple outputs. The proposal modifies the standard evolutionary strategy commonly adopted in CGP by combining the subgraphs of the best outputs of the parent and its offspring in order to generate a new fittest individual. The proposed crossover is applied to combinational logic circuits with multiple outputs, a parameter analysis is performed, and the results obtained are compared to those found by a baseline CGP and other techniques from the literature.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133499435","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00089
V. G. T. D. Costa, S. M. Mastelini, A. Carvalho, Sylvio Barbon Junior
Recently, several classification algorithms capable of dealing with potentially infinite data streams have been proposed. One of the main challenges of this task is to continuously update predictive models to address concept drifts without compromise their predictive performance. Moreover, the classification algorithm used must be able to efficiently deal with processing time and memory limitations. In the data stream mining literature, ensemble-based classification algorithms are a good alternative to satisfy the previous requirements. These algorithms combine multiple weak learner algorithms, e.g., the Very Fast Decision Tree (VFDT), to create a model with higher predictive performance. However, the memory costs of each weak learner are stacked in an ensemble, compromising the limited space requirements. To manage the trade-off between accuracy, memory space, and processing time, this paper proposes to use the Strict VFDT (SVFDT) algorithm as an alternative weak learner for ensemble solutions which is capable of reducing memory consumption without harming the predictive performance. This paper experimentally compares two traditional and three state-of-the-art ensembles using as weak learners the VFDT and SVFDT across thirteen benchmark datasets. According to the experimental results, the proposed algorithm can obtain a similar predictive performance with a significant economy of memory space.
{"title":"Making Data Stream Classification Tree-Based Ensembles Lighter","authors":"V. G. T. D. Costa, S. M. Mastelini, A. Carvalho, Sylvio Barbon Junior","doi":"10.1109/BRACIS.2018.00089","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00089","url":null,"abstract":"Recently, several classification algorithms capable of dealing with potentially infinite data streams have been proposed. One of the main challenges of this task is to continuously update predictive models to address concept drifts without compromise their predictive performance. Moreover, the classification algorithm used must be able to efficiently deal with processing time and memory limitations. In the data stream mining literature, ensemble-based classification algorithms are a good alternative to satisfy the previous requirements. These algorithms combine multiple weak learner algorithms, e.g., the Very Fast Decision Tree (VFDT), to create a model with higher predictive performance. However, the memory costs of each weak learner are stacked in an ensemble, compromising the limited space requirements. To manage the trade-off between accuracy, memory space, and processing time, this paper proposes to use the Strict VFDT (SVFDT) algorithm as an alternative weak learner for ensemble solutions which is capable of reducing memory consumption without harming the predictive performance. This paper experimentally compares two traditional and three state-of-the-art ensembles using as weak learners the VFDT and SVFDT across thirteen benchmark datasets. According to the experimental results, the proposed algorithm can obtain a similar predictive performance with a significant economy of memory space.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"71 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120968808","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 : 2018-10-01DOI: 10.1109/bracis.2018.00022
P. P. Rebouças Filho, Navar de Medeiros Mendonça e Nascimento, Shara Shami Araújo Alves, Samuel Luz Gomes, Cláudio Marques de Sá Medeiros
Wind energy is an excellent source of alternative energy to complement the Brazilian energy matrix. However, one of the significant challenges lies in managing this resource, due to its intermittent behavior. This study addresses the estimation of the electric power production of the wind park, so its management could be more efficient. A real data from one-year records of wind speed and power from a wind park installed in a wind farm in Ceará State, Brazil, is used. At first, we provide a study of Logistic versus Least Squares regression to model the wind turbine power curve. Then, a novel variant of the Least Square Support Regression is used to forecast wind speed on the site. The Logistic regression demonstrated to be more suitable for the task of regression, and the wind speed forecasting with three steps ahead provided lower error rates. Our approach represents a system based on data from both wind turbine power and speed to serve as a tool for helping energy selling issues and scheduling turbine maintenance on periods of time with low energy production in the wind park.
{"title":"Estimation of the Energy Production in a Wind Farm Using Regression Methods and Wind Speed Forecast","authors":"P. P. Rebouças Filho, Navar de Medeiros Mendonça e Nascimento, Shara Shami Araújo Alves, Samuel Luz Gomes, Cláudio Marques de Sá Medeiros","doi":"10.1109/bracis.2018.00022","DOIUrl":"https://doi.org/10.1109/bracis.2018.00022","url":null,"abstract":"Wind energy is an excellent source of alternative energy to complement the Brazilian energy matrix. However, one of the significant challenges lies in managing this resource, due to its intermittent behavior. This study addresses the estimation of the electric power production of the wind park, so its management could be more efficient. A real data from one-year records of wind speed and power from a wind park installed in a wind farm in Ceará State, Brazil, is used. At first, we provide a study of Logistic versus Least Squares regression to model the wind turbine power curve. Then, a novel variant of the Least Square Support Regression is used to forecast wind speed on the site. The Logistic regression demonstrated to be more suitable for the task of regression, and the wind speed forecasting with three steps ahead provided lower error rates. Our approach represents a system based on data from both wind turbine power and speed to serve as a tool for helping energy selling issues and scheduling turbine maintenance on periods of time with low energy production in the wind park.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115183348","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00076
G. E. Rodrigues, Wilson Estécio Marcílio Júnior, D. M. Eler
Dimensionality Reduction is a commonly used method to reduce the number of dimensions of data. In this work, we verified its influence in classification process using combinations of projection techniques as dimensionality reduction algorithms. We also used Naïve Bayes and SMO as classifiers.
{"title":"Data Classification: Dimensionality Reduction Using Combined and Non-combined Multidimensional Projection Techniques","authors":"G. E. Rodrigues, Wilson Estécio Marcílio Júnior, D. M. Eler","doi":"10.1109/BRACIS.2018.00076","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00076","url":null,"abstract":"Dimensionality Reduction is a commonly used method to reduce the number of dimensions of data. In this work, we verified its influence in classification process using combinations of projection techniques as dimensionality reduction algorithms. We also used Naïve Bayes and SMO as classifiers.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121376821","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 : 2018-10-01DOI: 10.1109/BRACIS.2018.00047
G. Souza, D. F. S. Santos, R. G. Pires, J. Papa, A. Marana
Biometrics has been increasingly used as a safe and convenient technique for people identification. Despite the higher security of biometric systems, criminals have already developed methods to circumvent them, being the presentation of fake biometric information to the input sensor (spoofing attack) the easiest way. Face is considered one of the most promising biometric traits for people identification, including in mobile devices. However, face recognition systems can be easily fooled, for instance, by presenting to the sensor a printed photograph, a 3D mask, or a video recorded from the face of a legal user. Recently, despite some CNNs (Convolutional Neural Networks) based approaches have achieved state-of-the-art results in face spoofing detection, in most of the cases the proposed architectures are very deep, being unsuitable for devices with hardware restrictions. In this work, we propose an efficient architecture for face spoofing detection based on a width-extended CNN, which we called wCNN. Each part of wCNN is trained, separately, in a given region of the face, then their outputs are combined in order to decide whether the face presented to the sensor is real or fake. The proposed approach, which learns deep local features from each facial region due to its width-wide architecture, presented better accuracy than state-of-the-art methods, including the well-referenced fine-tuned VGG-Face, while being much more efficient regarding hardware resources and processing time.
{"title":"Efficient Width-Extended Convolutional Neural Network for Robust Face Spoofing Detection","authors":"G. Souza, D. F. S. Santos, R. G. Pires, J. Papa, A. Marana","doi":"10.1109/BRACIS.2018.00047","DOIUrl":"https://doi.org/10.1109/BRACIS.2018.00047","url":null,"abstract":"Biometrics has been increasingly used as a safe and convenient technique for people identification. Despite the higher security of biometric systems, criminals have already developed methods to circumvent them, being the presentation of fake biometric information to the input sensor (spoofing attack) the easiest way. Face is considered one of the most promising biometric traits for people identification, including in mobile devices. However, face recognition systems can be easily fooled, for instance, by presenting to the sensor a printed photograph, a 3D mask, or a video recorded from the face of a legal user. Recently, despite some CNNs (Convolutional Neural Networks) based approaches have achieved state-of-the-art results in face spoofing detection, in most of the cases the proposed architectures are very deep, being unsuitable for devices with hardware restrictions. In this work, we propose an efficient architecture for face spoofing detection based on a width-extended CNN, which we called wCNN. Each part of wCNN is trained, separately, in a given region of the face, then their outputs are combined in order to decide whether the face presented to the sensor is real or fake. The proposed approach, which learns deep local features from each facial region due to its width-wide architecture, presented better accuracy than state-of-the-art methods, including the well-referenced fine-tuned VGG-Face, while being much more efficient regarding hardware resources and processing time.","PeriodicalId":405190,"journal":{"name":"2018 7th Brazilian Conference on Intelligent Systems (BRACIS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121516959","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：