Pub Date : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00066
Zirui Leng
With the development of deep learning in recent years, artificial intelligence has been widely applied in daily lives, industries, and services, which has attracted widespread attention. Based on the above application, this paper studies the typical application technology of artificial intelligence, and builds an “emotional intelligence” model using traditional facial emotion recognition as an example, accelerating the response of the model as much as possible while ensuring correct recognition.
{"title":"Research on Optimizing Facial Expression Recognition Based on Convolutional Neural Network","authors":"Zirui Leng","doi":"10.1109/CONF-SPML54095.2021.00066","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00066","url":null,"abstract":"With the development of deep learning in recent years, artificial intelligence has been widely applied in daily lives, industries, and services, which has attracted widespread attention. Based on the above application, this paper studies the typical application technology of artificial intelligence, and builds an “emotional intelligence” model using traditional facial emotion recognition as an example, accelerating the response of the model as much as possible while ensuring correct recognition.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122115262","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00054
Chenyu Gu
With the rapid development of the Internet, phishing websites now show the characteristics of short life cycle and low construction cost, which leads to a large amount of data brought by the detection of phishing websites for URL (uniform resource locator). It will also lead to increased retrieval time and decreased detection speed. In order to deal with diverse, complex and hidden phishing websites, this paper proposes a lightweight framework for detecting phishing websites. We first choose the faster Minhash signature to match URLs. On one hand, similarity detection is employed if the websites is suspicious. On the other hand, based on machine learning, the phishing websites can be finally determined by intention detection without similar sites.
{"title":"A Lightweight Phishing Website Detection Algorithm by Machine Learning","authors":"Chenyu Gu","doi":"10.1109/CONF-SPML54095.2021.00054","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00054","url":null,"abstract":"With the rapid development of the Internet, phishing websites now show the characteristics of short life cycle and low construction cost, which leads to a large amount of data brought by the detection of phishing websites for URL (uniform resource locator). It will also lead to increased retrieval time and decreased detection speed. In order to deal with diverse, complex and hidden phishing websites, this paper proposes a lightweight framework for detecting phishing websites. We first choose the faster Minhash signature to match URLs. On one hand, similarity detection is employed if the websites is suspicious. On the other hand, based on machine learning, the phishing websites can be finally determined by intention detection without similar sites.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128900608","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00029
Bohao Zhang
The diagnosis of COVID-19 has become a highly focused research area that captures researchers’ attention worldwide. Although the results of RT-PCR have been regarded as the golden standard for diagnosing COVID-19, CT-based diagnostic systems also have their unique advantages, attracting numerous researchers continuously into the area of developing deep learning-based diagnostic systems that utilize CT images. This paper is committed to presenting a comprehensive review, including current dynamics, generalized framework and useful resources. To capture the pattern of the developed methods, this paper introduces a generalized framework containing two stages: segmentation and classification. Furthermore, various valuable online resources have also been collected to provide more datasets, existing implementations of diagnostic systems, and commonly adopted evaluation metrics to researchers that are new to this area for their better adaptation and contribution to this meaningful, life-changing field.
{"title":"A Comprehensive Review of Deep Learning-Based COVID-19 Detection Mechanisms Using CT Images","authors":"Bohao Zhang","doi":"10.1109/CONF-SPML54095.2021.00029","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00029","url":null,"abstract":"The diagnosis of COVID-19 has become a highly focused research area that captures researchers’ attention worldwide. Although the results of RT-PCR have been regarded as the golden standard for diagnosing COVID-19, CT-based diagnostic systems also have their unique advantages, attracting numerous researchers continuously into the area of developing deep learning-based diagnostic systems that utilize CT images. This paper is committed to presenting a comprehensive review, including current dynamics, generalized framework and useful resources. To capture the pattern of the developed methods, this paper introduces a generalized framework containing two stages: segmentation and classification. Furthermore, various valuable online resources have also been collected to provide more datasets, existing implementations of diagnostic systems, and commonly adopted evaluation metrics to researchers that are new to this area for their better adaptation and contribution to this meaningful, life-changing field.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129663423","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00044
Yutao Li
In today’s large and complex data background, data needs to be properly interpreted and expressed in order to convey information more clearly. In this paper, a powerful visualization tool, Tableau is used to make visual analysis of online sales data of an American supermarket, the results can better understand the information of sales situation. This can better assist decision-making and provide decision support for the managers of the supermarket.
{"title":"Application of Tableau in Visual Analysis Data of a US Supermarket Sales","authors":"Yutao Li","doi":"10.1109/CONF-SPML54095.2021.00044","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00044","url":null,"abstract":"In today’s large and complex data background, data needs to be properly interpreted and expressed in order to convey information more clearly. In this paper, a powerful visualization tool, Tableau is used to make visual analysis of online sales data of an American supermarket, the results can better understand the information of sales situation. This can better assist decision-making and provide decision support for the managers of the supermarket.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121001686","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00050
Lv Qinyang
Obstacle avoidance path planning is a key technology for autonomous vehicles in identifying obstacles and avoiding obstacles, which is of great significance to the development of autonomous driving technology. This article gives an overview of traditional algorithms and intelligent algorithms related to obstacle avoidance path planning technology for autonomous vehicles, analyzes, compares and summarizes the advantages and disadvantages of each algorithm, and introduces their combined application. Comprehensively considering the advantages and disadvantages of using a single algorithm to plan obstacle avoidance paths in practical applications, it is found that a single algorithm shows drawbacks in a dynamic environment, such as poor computing power. Therefore, it is concluded that the use of multiple algorithms can make up for the shortcomings of a single algorithm, which has many advantages and will be focus of automatic obstacle avoidance research in the future.
{"title":"Research and Analysis of Obstacle Avoidance Path Planning for Autonomous Vehicles","authors":"Lv Qinyang","doi":"10.1109/CONF-SPML54095.2021.00050","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00050","url":null,"abstract":"Obstacle avoidance path planning is a key technology for autonomous vehicles in identifying obstacles and avoiding obstacles, which is of great significance to the development of autonomous driving technology. This article gives an overview of traditional algorithms and intelligent algorithms related to obstacle avoidance path planning technology for autonomous vehicles, analyzes, compares and summarizes the advantages and disadvantages of each algorithm, and introduces their combined application. Comprehensively considering the advantages and disadvantages of using a single algorithm to plan obstacle avoidance paths in practical applications, it is found that a single algorithm shows drawbacks in a dynamic environment, such as poor computing power. Therefore, it is concluded that the use of multiple algorithms can make up for the shortcomings of a single algorithm, which has many advantages and will be focus of automatic obstacle avoidance research in the future.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132920204","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00036
Shuying Shen
Accurate prediction of heart disease can save thousands of lives and de-crease health care cost significantly. In order to increase prediction accuracy-cy, we need to analyze data from multiple sources. However, current prediction methods based on machine learning do not consider the benefit of multiple sources. In this article, we combine four sensors with the electronic medical records (EMR), and perform feature extraction, preprocessing, feature fusion to predict heart disease by the support vector machines (SVM) and the convolutional neural network (CNN). The four sensors, including the medical sensor, the activity sensor, the sleeping sensor, and the emotion sensor use feature extraction techniques that are tailored for each sensor, considering their characteristics. Through analysis, it is demonstrated that the proposed method can increase the accuracy of heart disease prediction.
{"title":"A Multi-source Based Healthcare Method for Heart Disease Prediction by Machine Learning","authors":"Shuying Shen","doi":"10.1109/CONF-SPML54095.2021.00036","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00036","url":null,"abstract":"Accurate prediction of heart disease can save thousands of lives and de-crease health care cost significantly. In order to increase prediction accuracy-cy, we need to analyze data from multiple sources. However, current prediction methods based on machine learning do not consider the benefit of multiple sources. In this article, we combine four sensors with the electronic medical records (EMR), and perform feature extraction, preprocessing, feature fusion to predict heart disease by the support vector machines (SVM) and the convolutional neural network (CNN). The four sensors, including the medical sensor, the activity sensor, the sleeping sensor, and the emotion sensor use feature extraction techniques that are tailored for each sensor, considering their characteristics. Through analysis, it is demonstrated that the proposed method can increase the accuracy of heart disease prediction.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124160960","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00052
Peize Li
In recent years, deep learning models have achieved world-renowned achievements in the fields of image, speech and text recognition. However, the insufficient amount of labeled data has brought serious problems, and it is also difficult to identify unseen classes well. Therefore, if we want to achieve perfect recognition of unseen classes, we need to perform zero-shot learning. In order to solve the zero-shot learning problem, a better solution can be obtained by using the semantic space method. Zero-shot learning attempts to classify unseen data after learning the seen data. In this case, it is one of the most difficult learning methods to achieve perfect recognition. CLIP uses a data set of 400 million data pairs, resulting in higher efficiency and better robustness. Using the features obtained by traditional RESNET neural network and CLIP, two advanced methods, F-CLSWGAN and TF-VAEGAN, were tested. Through ZSL and GZSL experiments, excellent results have been achieved and the effectiveness of the combined method has been verified. This paper has tested the good effect of the application of CLIP on ZSL and GZSL. The experimental results show that CLIP has excellent performance on the AWA2 data set, whether it is using F-CLSWGAN or TF-VAEGAN. Among them, the effect of TF-VAEGAN is better.
{"title":"Application of CLIP on Advanced GAN of Zero-Shot Learning","authors":"Peize Li","doi":"10.1109/CONF-SPML54095.2021.00052","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00052","url":null,"abstract":"In recent years, deep learning models have achieved world-renowned achievements in the fields of image, speech and text recognition. However, the insufficient amount of labeled data has brought serious problems, and it is also difficult to identify unseen classes well. Therefore, if we want to achieve perfect recognition of unseen classes, we need to perform zero-shot learning. In order to solve the zero-shot learning problem, a better solution can be obtained by using the semantic space method. Zero-shot learning attempts to classify unseen data after learning the seen data. In this case, it is one of the most difficult learning methods to achieve perfect recognition. CLIP uses a data set of 400 million data pairs, resulting in higher efficiency and better robustness. Using the features obtained by traditional RESNET neural network and CLIP, two advanced methods, F-CLSWGAN and TF-VAEGAN, were tested. Through ZSL and GZSL experiments, excellent results have been achieved and the effectiveness of the combined method has been verified. This paper has tested the good effect of the application of CLIP on ZSL and GZSL. The experimental results show that CLIP has excellent performance on the AWA2 data set, whether it is using F-CLSWGAN or TF-VAEGAN. Among them, the effect of TF-VAEGAN is better.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128538500","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00034
Haoling Chen, Peng Liu
Stock return prediction has been a hot topic in both research and industry given its potential for large financial gain. The return signal, apart from its inherent volatility and complexity, is often accompanied by a multitude of noises, such as other stocks’ performance, macroeconomic factors and financial news, etc. To better characterize these factors, we propose a new model that consists of two levels of sequence: an NLP-based module to capture the sequential nature of words and sentences in the financial news, and a time-series-based module to exploit the sequential nature of adjacent observations in the stock price. In this proposed framework, we employ Hierarchical Attention Networks (HAN) in the text mining module, which could effectively model the financial news and extract important signals at both word and sentence level. For the time series module, the established Long-Short Term Memory (LSTM) network is used to model the complex serial dependence in the time series data. We compare with benchmark models using either module alone, as well as other alternatives using the traditional Bag of Words (BOW) approach, based on the Dow Jones Industrial Average (DJIA) dataset. Experiment results show that our proposal method performs better in several classification metrics for both positive and negative stock returns.
{"title":"Stock Return Prediction using Financial News: A Unified Sequence Model based on Hierarchical Attention and Long-Short Term Memory Networks","authors":"Haoling Chen, Peng Liu","doi":"10.1109/CONF-SPML54095.2021.00034","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00034","url":null,"abstract":"Stock return prediction has been a hot topic in both research and industry given its potential for large financial gain. The return signal, apart from its inherent volatility and complexity, is often accompanied by a multitude of noises, such as other stocks’ performance, macroeconomic factors and financial news, etc. To better characterize these factors, we propose a new model that consists of two levels of sequence: an NLP-based module to capture the sequential nature of words and sentences in the financial news, and a time-series-based module to exploit the sequential nature of adjacent observations in the stock price. In this proposed framework, we employ Hierarchical Attention Networks (HAN) in the text mining module, which could effectively model the financial news and extract important signals at both word and sentence level. For the time series module, the established Long-Short Term Memory (LSTM) network is used to model the complex serial dependence in the time series data. We compare with benchmark models using either module alone, as well as other alternatives using the traditional Bag of Words (BOW) approach, based on the Dow Jones Industrial Average (DJIA) dataset. Experiment results show that our proposal method performs better in several classification metrics for both positive and negative stock returns.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126662760","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00058
Y. Pan
As automation and informatization become a prevailing trend in industries, goods delivery and transportation are pivotal. Autonomous Guided Vehicles (AGV) is such a machine that can navigate while autonomously traveling. Besides saving labor, AGV can work in harsh and brutal conditions, which accounts for its popularity in industrial scenarios. In the following, we intend to discuss the design of the AGV navigation system from the perspective of vision parts. First, an in-depth comparison among different sensors on the existing navigation system will explain our choice of the visual navigation system. Then we will reveal three common challenges faced with visual navigation, i.e., poor illumination, limited view, and video data redundancy, and compare the merits and demerits of state-of-the-art solutions respectively. Our findings may offer practical suggestions for AGV design in real scenarios.
{"title":"Challenges in Visual Navigation of AGV and Comparison Study of Potential Solutions","authors":"Y. Pan","doi":"10.1109/CONF-SPML54095.2021.00058","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00058","url":null,"abstract":"As automation and informatization become a prevailing trend in industries, goods delivery and transportation are pivotal. Autonomous Guided Vehicles (AGV) is such a machine that can navigate while autonomously traveling. Besides saving labor, AGV can work in harsh and brutal conditions, which accounts for its popularity in industrial scenarios. In the following, we intend to discuss the design of the AGV navigation system from the perspective of vision parts. First, an in-depth comparison among different sensors on the existing navigation system will explain our choice of the visual navigation system. Then we will reveal three common challenges faced with visual navigation, i.e., poor illumination, limited view, and video data redundancy, and compare the merits and demerits of state-of-the-art solutions respectively. Our findings may offer practical suggestions for AGV design in real scenarios.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123441654","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 : 2021-11-01DOI: 10.1109/CONF-SPML54095.2021.00026
Binjun Jiang
Microblogging platforms are now one of the most popular means of social media in China. Carrying sentiment analysis on those platforms can provide valuable insights for various uses. However, the heavy use of Internet slang in microblog contexts and the lack of slang vocabulary in sentiment lexicons make it problematic. Aimed at this issue, we propose a method to build a comprehensive sentiment lexicon for Chinese internet slang. We leverage online sources to acquire a list of slang words first. Then, a method based on SO-PMI (Semantic Orientation from Pointwise Mutual Information) is used to assign the sentiment polarity to each word. By Utilizing online sources, the slang lexicon has comprehensive coverage of internet slang. The sentiment categorization method based on SO-PMI guarantees the sentiment polarity we acquire from microblog flatforms is compatible with the same microblog context the lexicon aimed to analyze.
{"title":"Building a Chinese Slang Sentiment Lexicon Using Online Crowdsourcing Dictionaries","authors":"Binjun Jiang","doi":"10.1109/CONF-SPML54095.2021.00026","DOIUrl":"https://doi.org/10.1109/CONF-SPML54095.2021.00026","url":null,"abstract":"Microblogging platforms are now one of the most popular means of social media in China. Carrying sentiment analysis on those platforms can provide valuable insights for various uses. However, the heavy use of Internet slang in microblog contexts and the lack of slang vocabulary in sentiment lexicons make it problematic. Aimed at this issue, we propose a method to build a comprehensive sentiment lexicon for Chinese internet slang. We leverage online sources to acquire a list of slang words first. Then, a method based on SO-PMI (Semantic Orientation from Pointwise Mutual Information) is used to assign the sentiment polarity to each word. By Utilizing online sources, the slang lexicon has comprehensive coverage of internet slang. The sentiment categorization method based on SO-PMI guarantees the sentiment polarity we acquire from microblog flatforms is compatible with the same microblog context the lexicon aimed to analyze.","PeriodicalId":415094,"journal":{"name":"2021 International Conference on Signal Processing and Machine Learning (CONF-SPML)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123153489","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：