Project Articles

2023 CCF Conference on Artificial Intelligence (CCFAI 2023)

Default Latest Most Read

Please wait a minute... For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails Select Missing value imputation algorithm using dual discriminator based on conditional generative adversarial imputation network Jia SU, Hong YU Journal of Computer Applications    2024, 44 (5): 1423-1427.   DOI: 10.11772/j.issn.1001-9081.2023050697 Abstract （136）   HTML （9）    PDF （872KB）（134）       Knowledge map    Save Various factors in the application may cause data loss and affect the analysis of subsequent tasks. Therefore， the imputation of missing data values in data sets is particularly important. Moreover， the accuracy of data imputation can significantly impact the analysis of subsequent tasks. Incorrect imputation data may introduce more severe bias in the analysis compared to missing data. A new missing value imputation algorithm named DDC-GAIN （Dual Discriminator based on Conditional Generation Adversarial Imputation Network） was introduced based on Conditional Generative Adversarial Imputation Network （C-GAIN） and dual discriminator， in which the primary discriminator was assisted by the auxiliary discriminator in assessing the validity of predicted values. In other words， the authenticity of the generated sample was judged by global sample information and the relationship between features was emphasized to estimate predicted values. Experimental results on four datasets show that， compared with five classical imputation algorithms， DDC-GAIN algorithm achieves the lowest Root Mean Square Error （RMSE） under the same conditions and with large sample size； when the missing rate is 15% on the Default credit card dataset， the RMSE of DDC-GAIN is 28.99% lower than that of the optimal comparison algorithm C-GAIN. This indicates that it is effective to utilize the auxiliary discriminator to support the primary discriminator in learning feature relationships. Table and Figures | Reference | Related Articles | Metrics Select Oversampling algorithm based on synthesizing minority class samples using relationship between features Mingzhu LEI, Hao WANG, Rong JIA, Lin BAI, Xiaoying PAN Journal of Computer Applications    2024, 44 (5): 1428-1436.   DOI: 10.11772/j.issn.1001-9081.2023050803 Abstract （162）   HTML （9）    PDF （1836KB）（78）       Knowledge map    Save The phenomenon of data imbalance is very common in real life. In order to improve the overall classification accuracy， classifiers often misclassify minority class at the cost. But in real life， the consequences of misclassifying minority class may be very serious. Considering that the traditional resampling algorithm ignores the relationship between the spatial distribution of data and the sample features of minority class， a new sampling algorithm SABRF （Sampling Algorithm Based on Relationship between Features） was proposed to generate a new sample set. The key distinguishing features of imbalanced dataset were preserved through Pareto-based multi-objective feature selection， and the relationships among key features of minority class samples were captured through XGBoost （eXtreme Gradient Boosting） regression model. In addition， considering the quality of newly generated samples， a new sample selection strategy was proposed to retain better samples. Experiments were conducted on six publicly available UCI datasets and one real post-orthopedic thrombus dataset. Experimental results show that the proposed algorithm has good performance on Area Under receiver operating characteristic Curve （AUC）， F1 score （F1_score） and Geometric Mean （G_mean）. In addition， when using the new samples selected by the sample selection strategy based on multi-index evaluation for classification， the classification result of imbalanced data is also the best， which verifies the effectiveness of the sample selection strategy. Table and Figures | Reference | Related Articles | Metrics Select EraseMTS： iterative active multivariable time series anomaly detection algorithm based on margin anomaly candidate set Fan MENG, Qunli YANG, Jing HUO, Xinkuan WANG Journal of Computer Applications    2024, 44 (5): 1458-1463.   DOI: 10.11772/j.issn.1001-9081.2023050726 Abstract （166）   HTML （10）    PDF （1234KB）（70）       Knowledge map    Save Unsupervised anomaly detection methods for Multivariable Time Series （MTS） have attracted wide attention due to their low labeling costs. However， traditional unsupervised anomaly detection methods are often based on two assumptions： 1） Independent and Identical Distribution （IID） assumption， i.e.， there is no dependency between samples and attributes of MTS. 2） High-purity starting assumption， i.e.， it is assumed that a completely normal time series should be used for training. The above assumptions are often difficult to satisfy in practical scenarios. To address this problem， an iterative active MTS anomaly detection algorithm based on margin anomaly candidate set （called EraseMTS） was proposed. Firstly， a multi-granularity representation learning method was utilized to capture the dependencies within subsequences and between subsequences， and then represent the original MTS. Secondly， a selection strategy was proposed to interact with experts based on margin anomaly candidate set， which was determined by the subsequence anomaly score and the uncertainty of its anomaly degree. Finally， an iterative subsequence weight update mechanism was designed to integrate the abnormal feedback information into the training process of the unsupervised anomaly detection model. The performance of the initial training model was continuously optimized through iteration. The proposed algorithm was verified in detection performance， scalability， and stability respectively on four datasets in UCR time series archive and one synthetic dataset. Experimental results show that the proposed algorithm can run effectively and stably. Table and Figures | Reference | Related Articles | Metrics Select Few-shot object detection via fusing multi-scale and attention mechanism Hongtian LI, Xinhao SHI, Weiguo PAN, Cheng XU, Bingxin XU, Jiazheng YUAN Journal of Computer Applications    2024, 44 (5): 1437-1444.   DOI: 10.11772/j.issn.1001-9081.2023050699 Abstract （260）   HTML （13）    PDF （2781KB）（318）       Knowledge map    Save The existing two-stage few-shot object detection methods based on fine-tuning are not sensitive to the features of new classes， which will cause misjudgment of new classes into base classes with high similarity to them， thus affecting the detection performance of the model. To address the above issue， a few-shot object detection algorithm that incorporates multi-scale and attention mechanism was proposed， namely MA-FSOD （Few-Shot Object Detection via fusing Multi-scale and Attention mechanism）. Firstly， grouped convolutions and large convolution kernels were used to extract more class-discriminative features in the backbone network， and Convolutional Block Attention Module （CBAM） was added to achieve adaptive feature augmentation. Then， a modified pyramid network was used to achieve multi-scale feature fusion， which enables Region Proposal Network （RPN） to accurately find Regions of Interest （RoI） and provide more abundant high-quality positive samples from multiple scales to the classification head. Finally， the cosine classification head was used for classification in the fine-tuning stage to reduce the intra-class variance. Compared with the Few-Shot object detection via Contrastive proposal Encoding （FSCE） algorithm on PASCAL-VOC 2007/2012 dataset， the MA-FSOD algorithm improved AP50 for new classes by 5.6 percentage points； and on the more challenging MSCOCO dataset， compared with Meta-Faster-RCNN， the APs corresponding to 10-shot and 30-shot were improved by 0.1 percentage points and 1.6 percentage points， respectively. Experimental results show that MA-FSOD can more effectively alleviate the misclassification problem and achieve higher accuracy in few-shot object detection than some mainstream few-shot object detection algorithms. Table and Figures | Reference | Related Articles | Metrics Select Two-stage search-based constrained evolutionary multitasking optimization algorithm Kaiwen ZHAO, Peng WANG, Xiangrong TONG Journal of Computer Applications    2024, 44 (5): 1415-1422.   DOI: 10.11772/j.issn.1001-9081.2023050696 Abstract （168）   HTML （11）    PDF （1756KB）（211）       Knowledge map    Save It is crucial in solving Constrained Multi-objective Optimization Problems （CMOPs） to efficiently balance the relationship between diversity， convergence and feasibility. However， the emergence of complex constraints poses a greater challenge in solving CMOPs. Therefore， a Two-stage search-based constrained Evolutionary Multitasking optimization Algorithm （TEMA） was proposed to achieve the balance between diversity， convergence and feasibility by completing the two cooperatively evolutionary tasks together. At first， the whole evolutionary process was divided into two stages， exploration stage and utilization stage， which were dedicated to enhance the extensive exploration capability and efficient search capability of the algorithm in the target space， respectively. Second， a dynamic constraint handling strategy was designed to balance the proportions of the feasible solutions in the population to enhance the exploration capability of the algorithm in the feasible region. Then， a backward search strategy was proposed to utilize the information contained in the unconstrained Pareto front to guide the algorithm to converge quickly to the constrained Pareto front. Finally， comparative experiments were performed on 23 problems in two benchmark test suites to verify the performance of the proposed algorithm. Experimental results indicate that the proposed algorithm achieves optimal IGD （Inverted Generational Distance） and HV （HyperVolume） values on 14 and 13 test problems， respectively， which reflects its significant advantages. Table and Figures | Reference | Related Articles | Metrics Select Weakly supervised video anomaly detection based on triplet-centered guidance Zimeng ZHU, Zhixin LI, Zhan HUAN, Ying CHEN, Jiuzhen LIANG Journal of Computer Applications    2024, 44 (5): 1452-1457.   DOI: 10.11772/j.issn.1001-9081.2023050748 Abstract （197）   HTML （10）    PDF （2177KB）（109）       Knowledge map    Save In view of the complex diversity and short time persistence of surveillance video anomaly， a weakly supervised video abnormal detection method was introduced to detect anomalies by only using video-level tags， and an anomaly regression network VLARNet based on Variational AutoEncoder （VAE） and Long Short-Term Memory （LSTM） network was proposed as an anomaly detection framework to effectively capture the temporal dependencies in time series data， eliminate redundant information and retain key information in the data. Anomaly detection was considered as a regression problem by VLARNet. To learn detection features， a Triplet-Centered Loss for Anomaly Score Regression （TCLASR） was designed and combined with Dynamic Multiple Instance Learning loss （DMIL） to further improve the discrimination ability of features. The DMIL widened the inter-class distance between abnormal instances and normal instances， but it also widened the intra-class distance. The TCLASR made the distances between the instances in the same class and the center closer and the distances between instances in different classes and the center farther. The proposed VLARNet was comprehensively tested on ShanghaiTech and CUHK Avenue datasets. Experimental results show that VLARNet can effectively utilize various information in video data， achieving Area Under receiver operating characteristic Curve （AUC） of 94.64% and 93.00% respectively on the two datasets， which is significantly better than those of the comparison algorithms. Table and Figures | Reference | Related Articles | Metrics Select Sleep stage classification model by meta transfer learning in few-shot scenarios Wangjun SHI, Jing WANG, Xiaojun NING, Youfang LIN Journal of Computer Applications    2024, 44 (5): 1445-1451.   DOI: 10.11772/j.issn.1001-9081.2023050747 Abstract （239）   HTML （11）    PDF （1546KB）（110）       Knowledge map    Save Sleep disorders are receiving more and more attention， and the accuracy and generalization of automated sleep stage classification are facing more and more challenges. However， due to the very limited human sleep data publicly available， the sleep stage classification task is actually similar to a few-shot scenario. And due to the widespread individual differences in sleep features， it is difficult for existing machine learning models to guarantee accurate classification of data from new subjects who have not participated in the training. In order to achieve accurate stage classification of new subjects’ sleep data， existing studies usually require additional collection and labeling of large amounts of data from new subjects and personalized fine-tuning of the model. Based on this， a new sleep stage classification model， Meta Transfer Sleep Learner （MTSL）， was proposed. Inspired by the idea of Scale & Shift based weight transfer strategy in transfer learning， a new meta transfer learning framework was designed. The training phase included two steps： pre-training and meta transfer training， and many meta-tasks were used for meta transfer training. In the test phase， the model could be easily adapted to the feature distribution of new subjects by fine-tuning with only a few new subjects’ data， which greatly reduced the cost of accurate sleep stage classification for new subjects. Experimental results on two public sleep datasets show that MTSL model can achieve higher accuracy and F1-score under both single-dataset and cross-dataset conditions. This indicates that MTSL is more suitable for sleep stage classification tasks in few-shot scenarios. Table and Figures | Reference | Related Articles | Metrics