Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837754
Aman Kumar, Manish Khare, Saurabh Tiwari
The sentiments of developers play a major role in productivity, code quality, and satisfaction. The workload of the developers and their interest in a specific programming language affect the overall quality of the development process. Open source projects, where developers (or contributors) work based on their interest in contributing to the project apart of their routine work. In this paper, we are analysing the sentiments of the developers on GitHub while working on different open source projects. Our study mainly focuses on three aspects: (1) analysing the day of the week in which the comment was made by the developer, (2) emotions of the developer throughout the course of a project, and (3) emotions with different programming languages. The analysis was done by looking into the developer comments on issues, pull requests, and comments for the repository. Our results show that projects developed on Monday’s tend to more negative emotion. Additionally, comments written in issues have higher negative polarity in their sentimental content, and projects developed in Java and Python have more positive comments as compared to C and C++.
{"title":"Sentiment Analysis of Developers’ Comments on GitHub Repository: A Study","authors":"Aman Kumar, Manish Khare, Saurabh Tiwari","doi":"10.1109/icaci55529.2022.9837754","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837754","url":null,"abstract":"The sentiments of developers play a major role in productivity, code quality, and satisfaction. The workload of the developers and their interest in a specific programming language affect the overall quality of the development process. Open source projects, where developers (or contributors) work based on their interest in contributing to the project apart of their routine work. In this paper, we are analysing the sentiments of the developers on GitHub while working on different open source projects. Our study mainly focuses on three aspects: (1) analysing the day of the week in which the comment was made by the developer, (2) emotions of the developer throughout the course of a project, and (3) emotions with different programming languages. The analysis was done by looking into the developer comments on issues, pull requests, and comments for the repository. Our results show that projects developed on Monday’s tend to more negative emotion. Additionally, comments written in issues have higher negative polarity in their sentimental content, and projects developed in Java and Python have more positive comments as compared to C and C++.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115340799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837594
Na Wu, Zongwu Ke, Lei Feng
The prediction of time series data is very difficult. For example, the price of stocks belongs to time series. Small fluctuations in society, politics, economy and culture may affect the stocks in the stock market. In the stock market, it is very important for people to have a general judgment on stocks. Therefore, the study of stocks has practical significance. This experiment confirms that the results are affected by the data set and statesize. Statesize is predicted by the closing price of several days.On the premise that the appropriate size of statesize makes the final profit the highest, and on the premise that improved algorithm of Q value based on DQN adds regularization (DDQN), it is proved that under different data sets, adding Long Short-Term Memory (LSTM) and full connection layer are better than only full connection layer. DQN is composed of neural network and Q-learning. Q-learning is a basic algorithm in reinforcement learning. And it is proved that DDQN algorithm is better than DQN on the premise that the appropriate statesize makes the final profit the highest, and on the premise of adding regularization and LSTM. Finally, it is also proved that under certain preconditions, the combination of LSTM and DDQN is better than only DQN and full connection layer. The only indicator of this experiment is the total profit. At the same time, this paper uses the closing price to predict.
{"title":"Stock Price Forecast Based on LSTM and DDQN","authors":"Na Wu, Zongwu Ke, Lei Feng","doi":"10.1109/icaci55529.2022.9837594","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837594","url":null,"abstract":"The prediction of time series data is very difficult. For example, the price of stocks belongs to time series. Small fluctuations in society, politics, economy and culture may affect the stocks in the stock market. In the stock market, it is very important for people to have a general judgment on stocks. Therefore, the study of stocks has practical significance. This experiment confirms that the results are affected by the data set and statesize. Statesize is predicted by the closing price of several days.On the premise that the appropriate size of statesize makes the final profit the highest, and on the premise that improved algorithm of Q value based on DQN adds regularization (DDQN), it is proved that under different data sets, adding Long Short-Term Memory (LSTM) and full connection layer are better than only full connection layer. DQN is composed of neural network and Q-learning. Q-learning is a basic algorithm in reinforcement learning. And it is proved that DDQN algorithm is better than DQN on the premise that the appropriate statesize makes the final profit the highest, and on the premise of adding regularization and LSTM. Finally, it is also proved that under certain preconditions, the combination of LSTM and DDQN is better than only DQN and full connection layer. The only indicator of this experiment is the total profit. At the same time, this paper uses the closing price to predict.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"301 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124297178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837603
Changqing Long, Houping Dai, Guodong Zhang, Junhao Hu
This paper explores the finite-time synchronization issue of a class of delayed fuzzy neural networks (DFNNs) by constructing new Lyapunov functional. Under the novel adaptive controller, sufficient conditions are derived to assure the finite-time synchronization of the considered DFNNs. In addition, the fuzzy logics are taken into accounted in the proposed network model, which complements and extends some of the existing results where the fuzzy logics or time delays are not considered. In the end, the validity of the derived synchronization results are verified by simulation examples.
{"title":"New Results on Finite-Time Synchronization of Delayed Fuzzy Neural Networks","authors":"Changqing Long, Houping Dai, Guodong Zhang, Junhao Hu","doi":"10.1109/icaci55529.2022.9837603","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837603","url":null,"abstract":"This paper explores the finite-time synchronization issue of a class of delayed fuzzy neural networks (DFNNs) by constructing new Lyapunov functional. Under the novel adaptive controller, sufficient conditions are derived to assure the finite-time synchronization of the considered DFNNs. In addition, the fuzzy logics are taken into accounted in the proposed network model, which complements and extends some of the existing results where the fuzzy logics or time delays are not considered. In the end, the validity of the derived synchronization results are verified by simulation examples.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114699839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837518
Honggang Yang, Rui Fan, Jiejie Chen, Mengfei Xu
In view of the possibility that Recurrent Neural Network(RNN)’s stochastic gradient descent method will converge to the local optimum problem, two fractional stochastic gradient descent methods are proposed in this paper. The methods respectively use the fractional order substitution derivative part defined by Caputo and the fractional order substitution difference form defined by Riemann Liouville to improve the updating method of network parameters. Combining with the gradient descent characteristics, the influence of fractional order on the training results is discussed, and two adaptive order adjustment methods are proposed. Experiments on MNIST and FashionMNIST datasets show that the fractional stochastic gradient optimization algorithm can improve the classification accuracy and training speed of recurrent neural network.
{"title":"Recurrent Neural Networks with Fractional Order Gradient Method","authors":"Honggang Yang, Rui Fan, Jiejie Chen, Mengfei Xu","doi":"10.1109/icaci55529.2022.9837518","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837518","url":null,"abstract":"In view of the possibility that Recurrent Neural Network(RNN)’s stochastic gradient descent method will converge to the local optimum problem, two fractional stochastic gradient descent methods are proposed in this paper. The methods respectively use the fractional order substitution derivative part defined by Caputo and the fractional order substitution difference form defined by Riemann Liouville to improve the updating method of network parameters. Combining with the gradient descent characteristics, the influence of fractional order on the training results is discussed, and two adaptive order adjustment methods are proposed. Experiments on MNIST and FashionMNIST datasets show that the fractional stochastic gradient optimization algorithm can improve the classification accuracy and training speed of recurrent neural network.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123972228","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}
This paper addresses the bipartite synchronization of coupled neural networks with time-varying delay. By introducing an effective quantized controller, the bipartite synchronization of coupled neural networks with time-varying delay is realized and sufficient conditions for assuring the bipartite synchronization are derived in virtue of a Halanay inequality. Moreover, the bipartite synchronization of coupled neural networks without delay via quantized controller is also taken into account in corollary as a special case. In the end, a numerical example is provided to demonstrate the correctness of theoretical results.
{"title":"Asymptotic Bipartite Synchronization of Coupled Neural Networks Via Quantized Control","authors":"Ting Liu, Junhong Zhao, Peng Liu, Jian Yong, Shulong Fan, Junwei Sun","doi":"10.1109/icaci55529.2022.9837729","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837729","url":null,"abstract":"This paper addresses the bipartite synchronization of coupled neural networks with time-varying delay. By introducing an effective quantized controller, the bipartite synchronization of coupled neural networks with time-varying delay is realized and sufficient conditions for assuring the bipartite synchronization are derived in virtue of a Halanay inequality. Moreover, the bipartite synchronization of coupled neural networks without delay via quantized controller is also taken into account in corollary as a special case. In the end, a numerical example is provided to demonstrate the correctness of theoretical results.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127202207","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}
Session-based recommendation mainly solves the recommendation problem in the anonymous scene, which is a challenging task. In recent years, most methods based on graph neural network (GNN) have ignore the location information of neighboring items. So we propose a graph aggregation method that introduces relative location information to capture this information. Specifically, we use two methods to learn item embedding, the location graph aggregation method is mainly used to capture the location relationship information between neighbors, and common graph aggregation method is mainly used to capture higher-order relationship information between items. Finally, we construct a session recommendation model and demonstrate the effectiveness of the proposed method on three datasets.
{"title":"An Improved Graph Neural Network Method Using Relative Position Information for Session-based Recommendation","authors":"Shuai Zhang, Yujie Xiao, Mingze Li, Xiaowei Li, Benhui Chen","doi":"10.1109/icaci55529.2022.9837599","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837599","url":null,"abstract":"Session-based recommendation mainly solves the recommendation problem in the anonymous scene, which is a challenging task. In recent years, most methods based on graph neural network (GNN) have ignore the location information of neighboring items. So we propose a graph aggregation method that introduces relative location information to capture this information. Specifically, we use two methods to learn item embedding, the location graph aggregation method is mainly used to capture the location relationship information between neighbors, and common graph aggregation method is mainly used to capture higher-order relationship information between items. Finally, we construct a session recommendation model and demonstrate the effectiveness of the proposed method on three datasets.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131389474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837669
Fei Wei, Guici Chen, Lei Yu
This paper investigates the finite-time anti-synchronization problem for a class of memristor oscillation circuit systems. Firstly, a 4th order memristor chaotic circuit system is derived using a quadratic nonlinear activated magneto-controlled memristor instead of the Chua diode in the Chua circuit. Then, a suitable controller is designed utilizing Lyapunov stability theory to drive the drive-response systems to finite-time anti-synchronization. Furthermore, the derived synchronization criterion is related to the system parameters. Therefore, the results obtained are more general and extend previous work. Finally, a numerical example is given and simulated to verify the validity of the results obtained.
{"title":"Finite-time Anti-synchronization of Memristor Oscillation System","authors":"Fei Wei, Guici Chen, Lei Yu","doi":"10.1109/icaci55529.2022.9837669","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837669","url":null,"abstract":"This paper investigates the finite-time anti-synchronization problem for a class of memristor oscillation circuit systems. Firstly, a 4th order memristor chaotic circuit system is derived using a quadratic nonlinear activated magneto-controlled memristor instead of the Chua diode in the Chua circuit. Then, a suitable controller is designed utilizing Lyapunov stability theory to drive the drive-response systems to finite-time anti-synchronization. Furthermore, the derived synchronization criterion is related to the system parameters. Therefore, the results obtained are more general and extend previous work. Finally, a numerical example is given and simulated to verify the validity of the results obtained.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121718719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837674
Hanwen Liu, Bingrong Xu, Yin Sheng, Zhigang Zeng
An interesting property of deep convolutional neural networks is their weakness to adversarial examples, which can deceive the models with subtle perturbations. Though adversarial attack algorithms have accomplished excellent performance in the white-box scenario, they frequently display a low attack success rate in the black-box scenario. Various transformation-based attack methods are shown to be powerful to enhance the transferability of adversarial examples. In this work, several novel transformation-based attack methods that integrate with the Random Block Shuffle (RBS) and Ensemble Random Block Shuffle (ERBS) mechanisms are come up with to boost adversarial transferability. First of all, the RBS calculates the gradient of the shuffled input instead of the original input. It increases the diversity of adversarial perturbation’s gradient and makes the original input’s information more invisible for the model. Based on the RBS, the ERBS is proposed to decrease gradient variance and stabilize the update direction further, which integrates the gradient of transformed inputs. Moreover, by incorporating various gradient-based attack methods with transformation-based methods, the adversarial transferability could be additionally improved fundamentally and relieve the overfitting problem. Our best attack method arrives an average success rate of 85.5% on two normally trained models and two adversarially trained models, which outperforms existing baselines.
{"title":"Boosting Adversarial Attack Transferability via Random Block Shuffle","authors":"Hanwen Liu, Bingrong Xu, Yin Sheng, Zhigang Zeng","doi":"10.1109/icaci55529.2022.9837674","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837674","url":null,"abstract":"An interesting property of deep convolutional neural networks is their weakness to adversarial examples, which can deceive the models with subtle perturbations. Though adversarial attack algorithms have accomplished excellent performance in the white-box scenario, they frequently display a low attack success rate in the black-box scenario. Various transformation-based attack methods are shown to be powerful to enhance the transferability of adversarial examples. In this work, several novel transformation-based attack methods that integrate with the Random Block Shuffle (RBS) and Ensemble Random Block Shuffle (ERBS) mechanisms are come up with to boost adversarial transferability. First of all, the RBS calculates the gradient of the shuffled input instead of the original input. It increases the diversity of adversarial perturbation’s gradient and makes the original input’s information more invisible for the model. Based on the RBS, the ERBS is proposed to decrease gradient variance and stabilize the update direction further, which integrates the gradient of transformed inputs. Moreover, by incorporating various gradient-based attack methods with transformation-based methods, the adversarial transferability could be additionally improved fundamentally and relieve the overfitting problem. Our best attack method arrives an average success rate of 85.5% on two normally trained models and two adversarially trained models, which outperforms existing baselines.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129441258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15DOI: 10.1109/icaci55529.2022.9837581
Zhansong Ma, Bingrong Xu, Lei Wang, Hanwen Liu, Zhigang Zeng
Unsupervised domain adaptation (UDA) recognizes unlabeled domain data by using the classifier learned from another domain. Previous works mainly focus on domain-level alignment that usually ignores the class-level information, resulting in the samples of different classes being too close to be classified correctly. To tackle this challenge, we design a unified weighted maximum mean discrepancy (MMD) metric method, that measures the differences in empirical distributions of two domains by calculating the weights of different sample pairs adaptively. The unified weighted MMD method is proposed which combines the class-level alignment with domain-level alignment, making full use of intra-domain, inter-domain, intra-class, and inter-class information with adaptive weights, and it is easy to implement. Experiment results demonstrate that our method can obtain superior results from two standard UDA datasets Office-31 and ImageCLEF-DA, compared with other UDA approaches.
{"title":"A Unified Weighted MMD For Unsupervised Domain Adaptation","authors":"Zhansong Ma, Bingrong Xu, Lei Wang, Hanwen Liu, Zhigang Zeng","doi":"10.1109/icaci55529.2022.9837581","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837581","url":null,"abstract":"Unsupervised domain adaptation (UDA) recognizes unlabeled domain data by using the classifier learned from another domain. Previous works mainly focus on domain-level alignment that usually ignores the class-level information, resulting in the samples of different classes being too close to be classified correctly. To tackle this challenge, we design a unified weighted maximum mean discrepancy (MMD) metric method, that measures the differences in empirical distributions of two domains by calculating the weights of different sample pairs adaptively. The unified weighted MMD method is proposed which combines the class-level alignment with domain-level alignment, making full use of intra-domain, inter-domain, intra-class, and inter-class information with adaptive weights, and it is easy to implement. Experiment results demonstrate that our method can obtain superior results from two standard UDA datasets Office-31 and ImageCLEF-DA, compared with other UDA approaches.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122457655","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}
The safety and reliability of the pantograph are critical and essential maintenance tasks in the railway transportation system. The majority of previous efforts proposed intelligent detection methods for achieving rapid and accurate inspection of the pantograph's health status. However, no research has been conducted on the automatic generation of pantograph health status reports, which is the primary reference basis for maintenance decisions. In this paper, in the light of the successful work of DenseCap, a pantograph image captioning model (PanCap for short) is proposed, which replaces VGG-16 with ResNet-50-FPN as the backbone to extract richer image features. In addition, Focal Loss and Transformer are used in PanCap to improve the description performance by addressing the problems of classification imbalance and dependent description. Evaluate the Visual Genome (VG) and pantograph image dataset, and the effectiveness of the proposed method is demonstrated by the experimental results.
{"title":"Automatic Pantograph Health Status Report Generation Based on Dense Captioning","authors":"Xinqiang Yin, Xiukun Wei, Zhaoxin Li, Dehua Wei, Qingfeng Tang","doi":"10.1109/icaci55529.2022.9837656","DOIUrl":"https://doi.org/10.1109/icaci55529.2022.9837656","url":null,"abstract":"The safety and reliability of the pantograph are critical and essential maintenance tasks in the railway transportation system. The majority of previous efforts proposed intelligent detection methods for achieving rapid and accurate inspection of the pantograph's health status. However, no research has been conducted on the automatic generation of pantograph health status reports, which is the primary reference basis for maintenance decisions. In this paper, in the light of the successful work of DenseCap, a pantograph image captioning model (PanCap for short) is proposed, which replaces VGG-16 with ResNet-50-FPN as the backbone to extract richer image features. In addition, Focal Loss and Transformer are used in PanCap to improve the description performance by addressing the problems of classification imbalance and dependent description. Evaluate the Visual Genome (VG) and pantograph image dataset, and the effectiveness of the proposed method is demonstrated by the experimental results.","PeriodicalId":412347,"journal":{"name":"2022 14th International Conference on Advanced Computational Intelligence (ICACI)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121151705","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：