作者机构:
[Wang, Wenshuo; Li, Zengyang; Wang, Sicheng; Mo, Ran] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.;[Wang, Wenshuo; Li, Zengyang; Wang, Sicheng; Mo, Ran] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.;[Liang, Peng; Li, Bing] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.;[Liang, Peng] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.
会议名称:
31st IEEE/ACM International Conference on Program Comprehension (ICPC)
会议时间:
MAY 15-16, 2023
会议地点:
Melbourne, AUSTRALIA
会议主办单位:
[Li, Zengyang;Wang, Sicheng;Wang, Wenshuo;Mo, Ran] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Li, Zengyang;Wang, Sicheng;Wang, Wenshuo;Mo, Ran] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.^[Liang, Peng;Li, Bing] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.
会议论文集名称:
International Conference on Program Comprehension
关键词:
Deep Learning Framework;Bug;Multiple-Programming-Language Software System;Empirical Study
摘要:
Deep learning frameworks (DLFs) have been playing an increasingly important role in this intelligence age since they act as a basic infrastructure for an increasingly wide range of AI-based applications. Meanwhile, as multi-programming-language (MPL) software systems, DLFs are inevitably suffering from bugs caused by the use of multiple programming languages (PLs). Hence, it is of paramount significance to understand the bugs (especially the bugs involving multiple PLs, i.e., MPL bugs) of DLFs, which can provide a foundation for preventing, detecting, and resolving bugs in the development of DLFs. To this end, we manually analyzed 1497 bugs in three MPL DLFs, namely MXNet, PyTorch, and TensorFlow. First, we classified bugs in these DLFs into 12 types (e.g., algorithm design bugs and memory bugs) according to their bug labels and characteristics. Second, we further explored the impacts of different bug types on the development of DLFs, and found that deployment bugs and memory bugs negatively impact the development of DLFs in different aspects the most. Third, we found that 28.6%, 31.4%, and 16.0% of bugs in MXNet, PyTorch, and TensorFlow are MPL bugs, respectively; the PL combination of Python and C/C++ is most used in fixing more than 92% MPL bugs in all DLFs. Finally, the code change complexity of MPL bug fixes is significantly greater than that of single-programming-language (SPL) bug fixes in all the three DLFs, while in PyTorch MPL bug fixes have longer open time and greater communication complexity than SPL bug fixes. These results provide insights for bug management in DLFs.
作者机构:
[Pi, Chenchen; Xie, W; Xie, Wei; Sun, Hao] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.;[Pi, Chenchen; Xie, W; Xie, Wei; Sun, Hao] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.;[Pi, Chenchen; Xie, W; Xie, Wei; Sun, Hao] Cent China Normal Univ, Natl Language Resources Monitoring & Res Ctr Netw, Wuhan, Peoples R China.
会议名称:
IEEE International Conference on Multimedia and Expo (ICME)
会议时间:
JUL 10-14, 2023
会议地点:
Brisbane, AUSTRALIA
会议主办单位:
[Sun, Hao;Pi, Chenchen;Xie, Wei] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.^[Sun, Hao;Pi, Chenchen;Xie, Wei] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.^[Sun, Hao;Pi, Chenchen;Xie, Wei] Cent China Normal Univ, Natl Language Resources Monitoring & Res Ctr Netw, Wuhan, Peoples R China.
会议论文集名称:
IEEE International Conference on Multimedia and Expo
摘要:
Pseudo-labels are popular in semi-supervised facial expression recognition. Recent methods usually exploit the confidence as the criterion for pseudo-label generation, and utilize the high-confidence pseudo-labels as the ground-truth for training. However, high confidence cannot guarantee the correctness of pseudo-labels. False pseudo-labels can weaken the feature discrimination and degrade recognition performance. In this paper, we propose a Critical Feature Refinement Network (CFRN) to alleviate the interference of false pseudo-labels on the model performance. Specially, a feature dropout module and a feature emphasis module are proposed to improve the feature discrimination of CFRN. Then, a mean-absolute error loss is further exploited to improve the robustness against false pseudo-labels. Experimental results on three challenging datasets RAF-DB, SFEW and Affectnet demonstrate that the proposed CFRN outperforms the state-of-the-art methods.
摘要:
Question answering over temporal knowledge graphs (TKGQA) has attracted great attentions in natural language processing community. One of the key challenges is how to effectively model the representations of questions and the candidate answers associated with timestamp constraints. Many existing methods attempt to learn temporal knowledge graph embedding for entities, relations and timestamps. However, these existing methods cannot effectively exploiting temporal knowledge graph embeddings to capture time intervals (e.g., "WWII" refers to 1939-1945) as well as temporal relation words (e.g., "first" and "last") appeared in complex questions, resulting in the sub-optimal results. In this paper, we propose a temporal-sensitive information for complex question answering (TSIQA) framework to tackle these problems. We employ two alternative approaches to augment questions embeddings with question-specific time interval information, which consists of specific start and end timestamps. We also present auxiliary contrastive learning to contrast the answer prediction and prior knowledge regarding time approximation for questions that only differ by the temporal relation words. To evaluate the effectiveness of our proposed method, we conduct the experiments on CRONQUESTION. The results show that our proposed model achieves better improvements over the state-of-the-art models that require multiple steps of reasoning.
作者:
Wang, Tong;Cui, Jianqun;Chang, Yanan;Huang, Feng;Yang, Yi
作者机构:
[Huang, Feng; Wang, Tong] Cent China Normal Univ, Coll Phys Sci & Technol, Wuhan, Peoples R China.;[Cui, Jianqun; Chang, Yanan] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.;[Yang, Yi] NE Illinois Univ, Dept Comp Sci, Chicago, IL USA.
会议名称:
18th IEEE International Conference on Mobility, Sensing and Networking (MSN)
会议时间:
DEC 14-16, 2022
会议地点:
ELECTR NETWORK
会议主办单位:
[Wang, Tong;Huang, Feng] Cent China Normal Univ, Coll Phys Sci & Technol, Wuhan, Peoples R China.^[Cui, Jianqun;Chang, Yanan] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Yang, Yi] NE Illinois Univ, Dept Comp Sci, Chicago, IL USA.
关键词:
DTNs;traffic light;probability of encountering;node state
摘要:
Delay-Tolerant Networks (DTNs), a supplementary means of communication network in extreme situations, have aroused wide attention from scholars. However, it is challenging to efficiently utilize DTNs since they have intermittent and high-latency characteristics. In the design of DTNs routing scheme, the selection of relay nodes takes on a great significance in efficient communication. However, existing research has either considered only one of the node features, or simply fused node attributes without fully using their potential correlations. If the above problems are not effectively solved, the propagation of messages between nodes will become blind, and a considerable number of caches will be occupied and wasted by invalid copies. To solve the above challenges, a novel routing, "Traffic Light Routing Based on Node State Awareness (TLRNSA)", is proposed for efficient communication. To be specific, the node's own state, the environmental state, and the historical encounter state are synthesized. The traffic value of the node is obtained based on the adaptive weight adjustment mechanism. The node is divided into three traffic light states, including red, green, and yellow, in accordance with the traffic value. Different routing strategies are developed for the above three states to enhance their performance. The results of the comprehensive experiments suggested that TLRNSA outperforms other state-of-theart algorithms in delivery rate and latency. Compared with the two classic algorithms and the two optimized algorithms, the proposed method increases the delivery rate by 109.1%, 84.12%, 5.09%, and 1.09%, respectively, it reduces the delay by 32.16%, 36.46%, 32.77%, and 6.77%, respectively.
期刊:
IEEE/ACM Transactions on Computational Biology and Bioinformatics,2022年19(3):1322-1333 ISSN:1545-5963
通讯作者:
Jiang, X.
作者机构:
[He, Tingting; Jiang, Xingpeng; Ma, Yingjun] Cent China Normal Univ, Sch Comp, Wuhan 430079, Hubei, Peoples R China.;[Ma, Yingjun] Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Hubei, Peoples R China.;[He, Tingting; Jiang, Xingpeng] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan 430079, Hubei, Peoples R China.;[Tan, Yuting] Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Peoples R China.;[Tan, Yuting] Hubei Key Lab Math Sci, Wuhan 430079, Hubei, Peoples R China.
通讯机构:
Central China Normal University, School of Computer, Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning, Hubei, Wuhan, China
会议名称:
18th Asia Pacific Bioinformatics Conference (APBC)
会议时间:
AUG 18-20, 2020
会议地点:
ELECTR NETWORK
会议主办单位:
[Ma, Yingjun;He, Tingting;Jiang, Xingpeng] Cent China Normal Univ, Sch Comp, Wuhan 430079, Hubei, Peoples R China.^[Ma, Yingjun] Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Hubei, Peoples R China.^[He, Tingting;Jiang, Xingpeng] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan 430079, Hubei, Peoples R China.^[Tan, Yuting] Cent China Normal Univ, Sch Math & Stat, Wuhan 430079, Peoples R China.^[Tan, Yuting] Hubei Key Lab Math Sci, Wuhan 430079, Hubei, Peoples R China.
摘要:
Infectious diseases are currently the most important and widespread health problem, and identifying viral infection mechanisms is critical for controlling diseases caused by highly infectious viruses. Because of the lack of non-interactive protein pairs and serious imbalance between positive and negative sample ratios, the supervised learning algorithm is not suitable for prediction. At the same time, due to the lack of information on viral proteins and significant dissimilarity in sequence, some ensemble learning models have poor generalization ability. In this paper, we propose a Sequence-Based Ensemble Learning (Seq-BEL) method to predict the potential virus-human PPIs. Specifically, based on the amino acid sequence of proteins and the currently known virus-human PPI network, Seq-BEL calculates various features and similarities of human proteins and viral proteins, and then combines these similarities and features to score the potential of virus-human PPIs. The computational results show that Seq-BEL achieves success in predicting potential virus-human PPIs and outperforms other state-of-the-art methods. More importantly, Seq-BEL also has good predictive performance for new human proteins and new viral proteins. In addition, the model has the advantages of strong robustness and good generalization ability, and can be used as an effective tool for virus-human PPI prediction.
作者机构:
[Li, Wanxin; Wang, Wei; Jin, Lianghao; Xie, Wei] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.;[Li, Wanxin; Wang, Wei; Jin, Lianghao; Xie, Wei] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.;[Tu, Zhigang] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan, Peoples R China.
会议名称:
IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) / IEEE World Congress on Computational Intelligence (IEEE WCCI) / International Joint Conference on Neural Networks (IJCNN) / IEEE Congress on Evolutionary Computation (IEEE CEC)
会议时间:
JUL 18-23, 2022
会议地点:
Padua, ITALY
会议主办单位:
[Li, Wanxin;Xie, Wei;Wang, Wei;Jin, Lianghao] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.^[Li, Wanxin;Xie, Wei;Wang, Wei;Jin, Lianghao] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.^[Tu, Zhigang] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan, Peoples R China.
会议论文集名称:
IEEE International Joint Conference on Neural Networks (IJCNN)
摘要:
Group activity recognition aims to identify group activities from the videos. Most of the previous methods focus on modeling between individuals (one-to- one), which ignores the fact that a single individual's behavior may be jointly determined by multiple individual behaviors (many-to-one). For this reason, we propose a Multi-Hyperedge Hypergraph (MHH) to capture high-order relationships between multiple people. Specifically, we build three different types of hyperedges on the hypergraph structure. Each hyperedge can accommodate the characteristics of multiple nodes to capture different types of high-order relationships between nodes. Then, we use the late fusion method to fuse the three features to further enhance the overall behavioral representation. Finally, we perform a series of experiments on two of the most widely used benchmarks in group activity recognition, which have proved the effectiveness of MHH. More importantly, as far as we know, this is the first case of using a hypergraph structure for group activity recognition.
作者机构:
[Li, Wanxin; Wang, Wei; Jin, Lianghao; Xie, Wei] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.;[Li, Wanxin; Wang, Wei; Jin, Lianghao; Xie, Wei] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.;[Tu, Zhigang] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan, Peoples R China.
会议名称:
IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) / IEEE World Congress on Computational Intelligence (IEEE WCCI) / International Joint Conference on Neural Networks (IJCNN) / IEEE Congress on Evolutionary Computation (IEEE CEC)
会议时间:
JUL 18-23, 2022
会议地点:
Padua, ITALY
会议主办单位:
[Wang, Wei;Xie, Wei;Li, Wanxin;Jin, Lianghao] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.^[Wang, Wei;Xie, Wei;Li, Wanxin;Jin, Lianghao] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.^[Tu, Zhigang] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan, Peoples R China.
会议论文集名称:
IEEE International Joint Conference on Neural Networks (IJCNN)
摘要:
In skeleton-based action recognition task, graph convolutional network has attracted widespread attention and achieved remarkable results. However, most of the current methods are performing graph convolution on the entire skeleton graph, ignoring the fact that people are composed of different body parts. In addition, previous work ignores the temporal and spatial independence and relevance of different parts. Thus, to solve these issues, we optimize the representation of the skeleton graph, graph convolution and temporal convolution respectively. In this work, we propose multi-part adaptive graph convolution (MPA-GC) to adaptively learn the topology of each part of the body and dynamically aggregate the relevance between them. Meanwhile, we add a multi-scale temporal convolution module to better obtain temporal dimension features. Ultimately, we develop a powerful graph convolutional network named MPA-GCN, and extensive experiments on two public large-scale datasets NTU-RGB+D and NTU-RGB+D120 demonstrate the effectiveness of our module, which outperforms state-of-the-art methods.
作者机构:
[Cui, Jianqun; Chang, Yanan; Chang, YN; Chen, Huanhuan; Gong, Shuang; Chen, Ziyi] Cent China Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.;[Yang, Yi] NE Illinois Univ, Dept Comp Sci, Chicago, IL USA.
会议名称:
IEEE 28th International Conference on Parallel and Distributed Systems (IEEE ICPADS)
会议时间:
JAN 10-12, 2023
会议地点:
Nanjing, PEOPLES R CHINA
会议主办单位:
[Cui, Jianqun;Gong, Shuang;Chang, Yanan;Chen, Ziyi;Chen, Huanhuan] Cent China Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.^[Yang, Yi] NE Illinois Univ, Dept Comp Sci, Chicago, IL USA.
会议论文集名称:
International Conference on Parallel and Distributed Systems - Proceedings
摘要:
In delay-tolerant networks(DTN), the timeliness of the node's social circle and encounter time between nodes have a different effect in designing router algorithms. Considering these factors, this paper proposes an improved spray and wait algorithm based on the node social tree (TNST). Specifically, we will first combine the node's own attributes with social ability to calculate the node's delivery capability value. Followed by it, we build and update the social node tree with the delivery capability for each node. Finally, we will predict the encounter time between nodes, which based on their motion information. The node will select the encounter node as the relay node if the message's destination node is in the node social tree. Otherwise, the node whose social tree with higher propagation capacity will be selected as the relay node. Simulation results show that the TNST algorithm improves the delivery rate and reduces network overhead.
作者机构:
[Wu, Haifang; Zhao, Weizhong; He, Tingting; Jiang, Xingpeng; Luo, Shujie] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan 430079, Hubei, Peoples R China.;[Wu, Haifang; Zhao, Weizhong; He, Tingting; Jiang, Xingpeng; Luo, Shujie] Cent China Normal Univ, Sch Comp, Wuhan 430079, Hubei, Peoples R China.;[Wu, Haifang; Zhao, Weizhong; He, Tingting; Jiang, Xingpeng; Luo, Shujie] Cent China Normal Univ, Natl Language Resources Monitoring & Res Ctr Netw, Wuhan 430079, Hubei, Peoples R China.;[Zhao, Weizhong] Guilin Univ Elect Technol, Guangxi Key Lab Trusted Software, Guilin 541004, Peoples R China.;[Zhao, Weizhong] Guangxi Normal Univ, Guangxi Key Lab Multisource Informat Min & Secur, Guilin 541004, Peoples R China.
会议名称:
26th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)
会议时间:
MAY 16-19, 2022
会议地点:
SW Jiaotong Univ, Chengdu, PEOPLES R CHINA
会议主办单位:
SW Jiaotong Univ
会议论文集名称:
Lecture Notes in Artificial Intelligence
关键词:
Protein interface prediction;Sequence information;Structure information;Hybrid attention mechanism
摘要:
Protein interface prediction is fundamental to understand the hidden principles of many living activities. Although many approaches to the task of protein interface prediction have been proposed, most of existing methods fail to make full use of the available sequence information and structure information. To address the challenge, we propose a deep learning-based end-to-end framework for protein interface prediction, in which a hybrid attention mechanism is utilized to take into account the semantic associations and complementary effect between both sequence and structure information. More specifically, a cross-modal attention is built to capture the semantic associations between sequence representations and structure representations for proteins. In addition, a type-level attention is introduced to model the different contributions of sequence and structure information for predicting protein interaction interface. Experimental results on three commonly used datasets demonstrate the effectiveness of the proposed method.
作者机构:
[Cui, Jianqun; Chang, Yanan; Chen, Huanhuan; Gong, Shuang; Chen, Ziyi] Cent China Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.;[Yi, Yang] NE Illinois Univ, Dept Comp Sci, Chicago, IL USA.
会议名称:
19th IEEE International Conference on Mobile Ad Hoc and Smart Systems (MASS)
会议时间:
OCT 20-22, 2022
会议地点:
Denver, CO
会议主办单位:
[Cui, Jianqun;Chen, Huanhuan;Chang, Yanan;Chen, Ziyi;Gong, Shuang] Cent China Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.^[Yi, Yang] NE Illinois Univ, Dept Comp Sci, Chicago, IL USA.
关键词:
Tolerant network;social circle nodes;social transmission value;non-social transmission value;message copies
摘要:
In delay tolerant networks(DTN), non-social attributes and social attributes relating to nodes have a different effect in designing router algorithms, we can combine these two different attributes to design algorithms. In our paper, we put forward an improved spray-and-wait routing algorithm (NSSAWRouter) based on social relationship between nodes in DTN, which mainly selects relay nodes by adopting two different transmission values and assigns message copies between two nodes dynamically in the spray stage. Specifically, we will first build social circle nodes for each node and record the dynamic social relationship between a node and its social circle nodes. Followed by it, we introduce a concept called social transmission value, which is used to estimate the likelihood of a node comprehensively transmitting a message to the special node that message should be delivered to when the social circle nodes of the node contain the special node. Similarly, we use non-social transmission value to select relay nodes when the two encountered nodes' social circle nodes both do not contain the special node, which consists of node activity and delivery predictability. Finally, we use the proportion of two nodes' non-social transmission value or social transmission value to assign message copies, making nodes that have greater ability to send messages to the special node carry more message copies. The results of the simulation experiment illustrate that the algorithm proposed can increase the delivery rate and keep a relatively lower overload in comparison with DirectDelivery, Epidemic, Prophet, Spray and Wait, MASSRouter.
作者机构:
[Zhu, Runjie] York Univ, Informat Retrieval & Knowledge Management Res Lab, Toronto, ON, Canada.;[Xie, Zhiwen] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.;[Zhou, Guangyou] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.
会议名称:
IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT)
会议时间:
NOV 17-20, 2022
会议地点:
ELECTR NETWORK
会议主办单位:
[Zhu, Runjie] York Univ, Informat Retrieval & Knowledge Management Res Lab, Toronto, ON, Canada.^[Xie, Zhiwen] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Zhou, Guangyou] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.
关键词:
Bioinformation;COVID-19;Natural Language Processing;Node Classification;Text Mining
摘要:
Recent studies in machine learning have demonstrated the effectiveness of applying graph neural networks (GNNs) to single-cell RNA sequencing (scRNA-seq) data to predict COVID-19 disease states. In this study, we propose a graph attention capsule network (GACapNet) which extracts and fuses Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) transcriptomic patterns to improve node classification performance on cells and genes. Significantly different from the existing GNN approaches, we innovatively incorporate a capsule layer with dynamic routing into our model architecture to combine and fuse gene features effectively and to allow those more prominent gene features present in the output. We evaluate our GACapNet model on two scRNA-seq datasets, and the experimental results show that our GACapNet model significantly outperforms state-of-the-art baseline models. Therefore, our study demonstrates the capability of advanced machine learning models to generate predictive features and evolutionary patterns of the SARS-CoV-2 pathogen, and the applicability of closing knowledge gaps in the pathogenesis and recovery of COVID-19.
会议名称:
29th IEEE/ACM International Conference on Program Comprehension (ICPC) / 18th IEEE/ACM International Conference on Mining Software Repositories (MSR)
会议时间:
MAY 22-30, 2021
会议地点:
ELECTR NETWORK
会议主办单位:
[Li, Zengyang;Qi, Xiaoxiao;Yu, Qinyi;Mo, Ran] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Li, Zengyang;Qi, Xiaoxiao;Yu, Qinyi;Mo, Ran] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.^[Liang, Peng] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Yang, Chen] IBO Technol Shenzhen Co Ltd, Shenzhen, Peoples R China.
会议论文集名称:
International Conference on Program Comprehension
摘要:
Modern software systems, such as Spark, are usually written in multiple programming languages (PLs). Besides benefiting from code reuse, such systems can also take advantages of specific PLs to implement certain features, to meet various quality needs, and to improve development efficiency. In this context, a change to such systems may need to modify source files written in different PLs. We define a multi-programming-language commit (MPLC) in a version control system (e.g., Git) as a commit that involves modified source files written in two or more PLs. To our knowledge, the phenomenon of MPLCs in software development has not been explored yet. In light of the potential impact of MPLCs on development difficulty and software quality, we performed an empirical study to understand the state of MPLCs, their change complexity, as well as their impact on open time of issues and bug proneness of source files in real-life software projects. By exploring the MPLCs in 20 non-trivial Apache projects with 205,994 commits, we obtained the following findings: (1) 9% of the commits from all the projects are MPLCs, and the proportion of MPLCs in 80% of the projects goes to a relatively stable level; (2) more than 90% of the MPLCs from all the projects involve source files written in two PLs; (3) the change complexity of MPLCs is significantly higher than that of non-MPLCs in all projects; (4) issues fixed in MPLCs take significantly longer to be resolved than issues fixed in non-MPLCs in 80% of the projects; and (5) source files that have been modified in MPLCs tend to be more bug-prone than source files that have never been modified in MPLCs. These findings provide practitioners with useful insights on the architecture design and quality management of software systems written in multiple PLs.
摘要:
In this paper, we propose a new model that combines reinforcement learning and adversarial training to exploit the data generated by distant supervision for named entity recognition. Our model can not only reduce the influence of noise in generated data, but also find more informative instances for training. In the pre-training stage of the model, in order to make full use of the data generated by distant supervision, we use reinforcement learning to select reliable instances to pre-train a classifier. In the training stage of the model, we introduce the adversarial training mechanism, which can not only find more reliable instances to enhance the ability of the classifier, but also use noise data to improve the ability of the model to resist noise. To evaluate the performance of the model, we conduct experiments on two public datasets, Species800 dataset in biology and EC dataset in e-commerce domain. The experimental results show that in Species800 dataset, the F1 score of our model is 1.68% higher than that of baseline, and in EC dataset, the F1 score of our model is 6.32% higher than that of baseline. Compared to the state of art models, our model can achieve comparable performance just using word2vec embedding.
作者机构:
[Zhong, Rui; Huang, Yansen; Yao, Wenjin; Wang, Rui] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.
会议名称:
IEEE International Symposium on Circuits and Systems (IEEE ISCAS)
会议时间:
MAY 22-28, 2021
会议地点:
Daegu, SOUTH KOREA
会议主办单位:
[Huang, Yansen;Zhong, Rui;Yao, Wenjin;Wang, Rui] Cent China Normal Univ, Sch Comp, Wuhan, Peoples R China.
会议论文集名称:
IEEE International Symposium on Circuits and Systems
关键词:
Video Summarization;Attention model;Bi-LSTM
摘要:
The high redundancy among keyframes is a critical issue for the existing summarizing methods in dealing with user-created videos. To address the critical issue, we present an unsupervised learning method, Spatial Attention Model guided Bi-directional Long Short-term Memory network (Bi-LSTM), on the combination of visual and semantic features. As for the visual feature, we design a Salient-Area-Size-based spatial attention model on the observation that humans tend to focus on sizable and moving objects in videos. Moreover, the Bi-LSTM network is leveraged to exploit the semantic feature. Afterward, the Soft Selected Probability generated from the spatial attention and semantic feature is fused to obtain the final probability for keyframe selection. The reinforcement learning framework, trained by the Deep Deterministic Policy Gradient algorithm, is adopted to do unsupervised training. Extensive experiments on the SumMe and TVSum datasets demonstrate that our method outperforms the state-of-the-art methods in terms of F-score.
摘要:
The aim of multi-objective particle swarm optimizer (MOPSO) is to find an accurate and well-distributed approximation of the true Pareto Front (PF). The intrinsic character of PSO puts convergence first, which can cause great loss of population diversity. How to maintain the convergence and diversity simultaneously is an essential issue for MOPSO. In this paper, we propose a niche based multi-objective particle swarm optimizer (NMOPSO) to balance the convergence and diversity. First, a niche based on the Euclidean distance is constructed for each particle, then the leading particle is chosen out either from the niche or from the whole swarm. After that, two position update strategies are designed to update the position of each particle. The position update strategies provide two guiding models for leaders, one is utilizing the difference vector between the leader and the current particle, the other is directly taking some components of leaders. Three well-known test suites are employed to verify the performance of NMOPSO. Compared with three popular MOPSOs, simulation results show that NMOPSO performs better on most of test problems.
摘要:
Recently, Onscreen Marking (OSM) system based on the traditional cloud platform has been widely used in various large-scale public examinations. However, the mainstream examination marking process is not transparent, and there is the possibility of a black-box operation, which damages the fairness of the examination. Also, the issues related to data security and privacy are still considered to be serious challenges. In this paper, we deal with the above problems by providing secure and private transactions in a distributed OSM assuming the semi-trusted examination center. We have implemented a proof-of-concept for a consortium blockchain-based OSM in a privacy-preserving and auditable manner, enabling markers to anonymously mark to the distributed ledger.
作者机构:
[Li, Zengyang; Mo, Ran; Yu, Qinyi] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.;[Li, Zengyang; Mo, Ran; Yu, Qinyi] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.;[Liang, Peng] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.;[Yang, Chen] IBO Technol Shenzhen Co Ltd, Shenzhen, Peoples R China.
会议名称:
36th IEEE International Conference on Software Maintenance and Evolution (ICSME)
会议时间:
SEP 27-OCT 03, 2020
会议地点:
ELECTR NETWORK
会议主办单位:
[Li, Zengyang;Yu, Qinyi;Mo, Ran] Cent China Normal Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Li, Zengyang;Yu, Qinyi;Mo, Ran] Cent China Normal Univ, Hubei Prov Key Lab Artificial Intelligence & Smar, Wuhan, Peoples R China.^[Liang, Peng] Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China.^[Yang, Chen] IBO Technol Shenzhen Co Ltd, Shenzhen, Peoples R China.
会议论文集名称:
Proceedings-IEEE International Conference on Software Maintenance
关键词:
defect technical debt;interest;interest measure;case study
摘要:
Defect technical debt (defect debt) refers to known deferred bugs that have not been fixed. The interest of a defect debt item (i.e., bug) is the extra effort needed to fix the bug due to the delay of fixing. It is important to measure defect debt interest in a software system in order to be aware of how much the interest of certain bugs is and which bugs should be fixed first. Furthermore, it is valuable to understand the features of the bugs of high interest or no interest, so as to facilitate the identification of those kinds of bugs. In this work, we proposed three pairs of measures for quantifying defect debt interest at three granularities (i.e., lines of code, source file, and package) of software changes, and conducted an exploratory case study on 13,438 bugs collected from 59 non-trivial Apache open source software projects written mainly in Java. We have the following findings. (1) Each bug, with an average of 224 days delay of fixing, has interest of 660 to 845 lines of code, 0.57 to 1.05 source files, and 0.29 to 0.35 package on average in need for understanding or modifying when fixing the bug. (2) The average interest of a bug shows overall increasing trends over delayed time of bug fixing and bug priority levels from the lowest to highest. (3) Around 30% of the bugs under study did not incur interest at the granularity of lines of code, which means that the involved source files in the bug-fixing commits of such bugs were not modified at all during the delayed time of bug fixing. (4) The average code change size of bug-fixing commits of a bug without interest is much smaller than that of a bug with (high) interest; for a bug without interest at the granularity of lines of code, the source files involved in the bug-fixing commits of the bug were alternately modified by different committers much less frequently than those of the bug with (high) interest.
会议名称:
2nd International Conference on Security with Intelligent Computing and Big-data Services (SICBS)
会议时间:
DEC 14-16, 2018
会议地点:
Guilin, PEOPLES R CHINA
会议主办单位:
[Xia, Zhe;Yang, Zhen;Xiong, Shengwu] Wuhan Univ Technol, Sch Comp Sci, Wuhan 430071, Hubei, Peoples R China.^[Hsu, Ching-Fang] Cent China Normal Univ, Comp Sch, Wuhan 430079, Hubei, Peoples R China.
会议论文集名称:
Advances in Intelligent Systems and Computing
摘要:
Secret sharing schemes allow the secret to be shared among a group of parties, so that a quorum of these parties can work together to recover the secret, but less number of parties cannot learn any information of the secret. In the literature, secret sharing schemes are normally analysed using heuristic arguments rather than strict security proofs. However, such a method may overlook some security flaws, especially when it is used to analyse the secrecy property. In this paper, we illustrate this issue using some concrete examples. We show that in two existing secret sharing schemes, the secrecy property was originally conjectured to be satisfied, but the adversary still can employ some security flaws to violate this property. We then introduce a game-based model that can be used to formally analyse the secrecy property in secret sharing schemes. We prove that our model captures the definition of the secrecy property. And as an example, we show how our method can be used to analyse Shamir secret sharing scheme. Note that our method might find applications in other secret sharing schemes as well.<br/> ©2020, Springer Nature Switzerland AG.