检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Wu, Xuan Liu, Zhangdaihong Zhao, Yizheng Yang, Yang Clifton, David A. The Oxford Suzhou Centre for Advanced Research Suzhou215123 China The Department of Engineering Science Institute of Biomedical Engineering University of Oxford OxfordOX1 2JD United Kingdom The National Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China The School of Public Health Shanghai Jiao Tong University School of Medicine Shanghai200025 China The Department of Engineering Science Institute of Biomedical Engineering University of Oxford OxfordOX1 2JD United Kingdom

Electronic health records (EHR) contain a comprehensive history of patients’ diagnostic, procedural, and prescription data, which are expressed in medical concepts coded in diverse standards, such as ICD-10-CM, ICD-10-PCS, and NDC. These medical concepts possess complex semantics e.g., causal/mutual exclusive/complication relationship between diagnoses concepts, which are implicitly embedded in EHR data. The semantics of medical concepts summarise medical context, which plays an important role in improving clinical tasks in the healthcare domain. Many EHR mining researchers utilize natural language processing (NLP) models for learning to represent medical concepts. However, there has been no systematic comparison of these models in representing medical concepts. In this study, we evaluated four NLP models, including Latent Dirichlet Allocation (LDA), Word2Vec, GloVe, and BERT, for medical concept representation learning using two large public EHR datasets in the pre-training setting. We discovered that GloVe performed the best among the four models. On the whole, medical concept representation learning models should incorporate global co-occurrence and subtle modelling for accurately capturing the relationships between code embeddings. © 2023, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Context decoupling augmentation for weakly supervised semantic segmentation

arXiv

引用

arXiv 2021年

作者： Su, Yukun Sun, Ruizhou Lin, Guosheng Wu, Qingyao School of Software and Engineering South China University of Technology China Key Laboratory of Big Data and Intelligent Robot Ministry of Education India School of Computer Science and Engineering Nanyang Technological University Singapore

Data augmentation is vital for deep learning neural networks. By providing massive training samples, it helps to improve the generalization ability of the model. Weakly supervised semantic segmentation (WSSS) is a challenging problem that has been deeply studied in recent years, conventional data augmentation approaches for WSSS usually employ geometrical transformations, random cropping and color jittering. However, merely increasing the same contextual semantic data does not bring much gain to the networks to distinguish the objects, e.g., the correct image-level classification of "aeroplane" may be not only due to the recognition of the object itself, but also its co-occurrence context like "sky", which will cause the model to focus less on the object features. To this end, we present a Context Decoupling Augmentation (CDA) method, to change the inherent context in which the objects appear and thus drive the network to remove the dependence between object instances and contextual information. To validate the effectiveness of the proposed method, extensive experiments on PASCAL VOC 2012 dataset with several alternative network architectures demonstrate that CDA can boost various popular WSSS methods to the new state-of-the-art by a large margin. Copyright © 2021, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Using evolutionary process for cross-version software defect prediction

引用

International Journal of Performability Engineering 2019年第9期15卷 2484-2493页

作者： Li, Yong Liu, Zhandong Zhang, Haijun College of Computer Science and Technology Xinjiang Normal University Urumqi830054 China Data Security Key Laboratory Xinjiang Normal University Urumqi830054 China Key Laboratory of Safety-Critical Software Ministry of Industry and Information Technology Nanjing211106 China

Cross-version defect prediction can effectively reduce the construction cost of a model, which is of great significance for understanding the causes of defects in subsequent software versions and improving the quality of software products. The objective of this work is to validate the feasibility of the cross-version defect predictor using the evolutionary process in different scenarios and to investigate practical guidelines for the choice of evolutionary process and historical data of a given project. Firstly, we verified the effectiveness of the evolutionary process attributes in the cross-version prediction model. Then, we analyzed the cross-version model prediction performance based on multi-version historical data. Finally, multiple benchmark models were selected for experimental comparison to further verify the effectiveness of the proposed method. The experimental conclusions can not only effectively reduce the construction cost of the defect prediction model in the version evolution scenarios, but also help developers understand the defect generation mechanism in the subsequent versions and improve the software development process in a targeted way. © 2019 Totem Publisher, Inc. All rights reserved.

关键词： Defects

来源：评论

学校读者我要写书评

暂无评论

ClickGuard: Exposing Hidden Click Fraud via Mobile Sensor Side-channel Analysis

ClickGuard: Exposing Hidden Click Fraud via Mobile Sensor Si...

引用

IEEE International Conference on Communications (ICC)

作者： Congcong Shi Rui Song Xinyu Qi Yubo Song Bin Xiao Sanglu Lu State Key Laboratory for Novel Software Technology Nanjing University School of Cyber Science and Engineering Southeast University Department of Computing The Hong Kong Polytechnic University

ISBN: (数字)9781728150895

ISBN: (纸本)9781728150901

Advertising income depends on the amount of clicks by users of websites and mobile applications. However, the emergence of click fraud greatly reduces the real benefits of the advertisement. Most existing researches focus on detecting click fraud by analyzing properties and patterns of click data streams, but attackers can construct data that looks legitimate by replaying former data streams. In this paper, we propose a novel system called ClickGuard to detect click fraud attacks. ClickGuard takes advantage of motion sensor signals from mobile devices, since the pattern of motion signals is completely different under real click events and fraud events. To prevent attackers from bypassing the system by faking the time-domain statistical characteristics of original signals, we introduce the MFCC algorithm in feature extraction phase. MFCC algorithm can extract frequency-domain features of original signals in specific frequency bands which are hardly constructed out of thin air. Classifiers are finally constructed using these features and several machine learning algorithms. Experiments show that ClickGuard can achieve the accuracy of 96.71% in general environment and 84.16% when attackers modify the time-domain statistical characteristics of raw data.

关键词： Feature extraction Frequency-domain analysis Machine learning algorithms Time-domain analysis Mel frequency cepstral coefficient IP networks Mobile handsets

来源：评论

学校读者我要写书评

暂无评论

Opcode sequence analysis of Android malware by a convolutional neural network

Opcode sequence analysis of Android malware by a convolution...

引用

作者： Li, Dan Zhao, Lichao Cheng, Qingfeng Lu, Ning Shi, Wenbo School of Computer Science and Engineering Northeastern University Shenyang China State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou China School of Computer Science and Technology Xidian University Xi'an China School of Computer and Communication Engineering Northeastern University at Qinhuangdao Qinhuangdao China

The number of malware has exploded due to the openness of the Android platform, and the endless stream of malware poses a threat to the privacy, tariffs, and device of mobile phone users. A novel Android mobile malware detection system is proposed, which employs an optimized deep convolutional neural network to learn from opcode sequences. The optimized convolutional neural network is trained multiple times by the raw opcode sequences extracted from the decompiled Android file, so that the feature information can be effectively learned and the malicious program can be detected more accurately. More critically, the k-max pooling method with better results is adopted in the pooling operation phase, which improves the detection effect of the proposed method. The experimental results show that the detection system achieved the accuracy of 99%, which is 2%-11% higher than the accuracy of the machine learning detection algorithms when using the same data set. It also ensures that the indicators, such as F1-score, recall, and precision, are maintained above 97%. Based on the detection system, a multi–data set comparison experiment is carried out. The introduced k-max pooling is deeply studied, and the effect of k of k-max pooling on the overall detection effect is observed. © 2019 John Wiley & Sons, Ltd.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

Plagiarism detection of multi-threaded programs using frequent behavioral pattern mining 32

Plagiarism detection of multi-threaded programs using freque...

引用

32nd International Conference on software Engineering and Knowledge Engineering, SEKE 2020

作者： Wang, Qing Tian, Zhenzhou Gao, Cong Chen, Lingwei School of Computer Science and Technology Xi'an University of Posts and Telecommunications Xi'an China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an China College of Information Sciences and Technology Pennsylvania State University PA United States

ISBN: (纸本)1891706500

software dynamic birthmark techniques construct birthmarks using the captured execution traces from running the programs, which serve as one of the most promising methods for obfuscation-resilient software plagiarism detection. However, due to the perturbation caused by non-deterministic thread scheduling in multi-threaded programs, such dynamic approaches optimized for sequential programs may suffer from the randomness in multi-threaded program plagiarism detection. In this paper, we propose a new dynamic thread-aware birthmark FPBirth to facilitate multi-threaded program plagiarism detection. We first explore dynamic monitoring to capture multiple execution traces with respect to system calls for each multi-threaded program under a specified input, and then leverage Apriori algorithm to mine frequent patterns to formulate our dynamic birthmark, which can not only depict the program's behavioral semantics, but also resist the changes and perturbations over execution traces caused by the thread scheduling in multithreaded programs. Using FPBirth, we design a multi-threaded program plagiarism detection system. The experimental results based on a public software plagiarism sample set demonstrate that the developed system integrating our proposed birthmark FPBirth cope better with multi-threaded plagiarism detection than alternative approaches. © 2020 Knowledge Systems Institute Graduate School. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A Method of Calculating the Semantic Similarity Between English and Chinese Concepts 4th

A Method of Calculating the Semantic Similarity Between Engl...

引用

4th International Conference on Machine Learning and Intelligent Communications, MLICOM 2019

作者： Cao, Jingwen Wang, Tiexin Li, Wenxin Tao, Chuanqi College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics 29# Jiangjun Road Jiangning District Nanjing211106 China Key Laboratory of Safety-Critical Software Nanjing University of Aeronautics and Astronautics Ministry of Industry and Information Technology Nanjing China State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

ISBN: (纸本)9783030323875

In the big data era, data and information processing is a common concern of diverse fields. To achieve the two keys "efficiency" and "intelligence" to the processing process, it’s necessary to search, define and build the potential links among heterogeneous data. Focusing on this issue, this paper proposes a knowledge-driven method to calculate the semantic similarity between (bilingual English-Chinese) words. This method is built on the knowledge base "HowNet", which defines and maintains the "atom taxonomy tree" and the "semantic dictionary" - a network of knowledge system describing the relationships between word concepts and attributes of the concepts. Compared to other knowledge bases, HowNet pays more attention to the connections between words based on concepts. Besides, this method is more complete in the analysis of concepts and more convenient in calculation methods. The non-relational database MongoDB is employed to improve the efficiency and fully use the rich knowledge maintained in HowNet. Considering both the structure of HowNet and characteristics of MongoDB, a certain number of equations are defined to calculate the semantic similarity. © 2019, ICST Institute for computer sciences, Social Informatics and Telecommunications Engineering.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Will Dependency Conflicts Affect My Program’s Semantics?

arXiv

引用

arXiv 2020年

作者： Wang, Ying Wu, Rongxin Wang, Chao Wen, Ming Liu, Yepang Cheung, ShingChi Yu, Hai Xu, Chang Zhu, Zhiliang Software College Northeasthern University China Department of Cyber Space Security Xiamen University China School of Cyber Science and Engineering Huazhong University of Science and Technology China Department of Computer Science and Engineering Southern University of Science and Technology China Department of Computer Science and Engineering Hong Kong University of Science and Technology China State Key Lab for Novel Software Technology and Department of Computer Science and Technology Nanjing University China

Java projects are often built on top of various third-party libraries. If multiple versions of a library exist on the classpath, JVM will only load one version and shadow the others, which we refer to as dependency conflicts. This would give rise to semantic conflict (SC) issues, if the library APIs referenced by a project have identical method signatures but inconsistent semantics across the loaded and shadowed versions of libraries. SC issues are difficult for developers to diagnose in practice, since understanding them typically requires domain knowledge. Although adapting the existing test generation technique for dependency conflict issues, RIDDLE, to detect SC issues is feasible, its effectiveness is greatly compromised. This is mainly because RIDDLE randomly generates test inputs, while the SC issues typically require specific arguments in the tests to be exposed. To address that, we conducted an empirical study of 75 real SC issues to understand the characteristics of such specific arguments in the test cases that can capture the SC issues. Inspired by our empirical findings, we propose an automated testing technique SENSOR, which synthesizes test cases using ingredients from the project under test to trigger inconsistent behaviors of the APIs with the same signatures in conflicting library versions. Our evaluation results show that SENSOR is effective and useful: it achieved a Precision of 0.803 and a Recall of 0.760 on open-source projects and a Precision of 0.821 on industrial projects;it detected 150 semantic conflict issues in 29 projects, 81.8% of which had been confirmed as real bugs. Copyright © 2020, The Authors. All rights reserved.

关键词： Libraries

来源：评论

学校读者我要写书评

暂无评论

Nearest neighborhood-based deep clustering for source data-absent unsupervised domain adaptation

arXiv

引用

arXiv 2021年

作者： Tang, Song Yang, Yan Ma, Zhiyuan Hendrich, Norman Zeng, Fanyu Ge, Shuzhi Sam Zhang, Changshui Zhang, Jianwei The Institute of Machine Intelligence University of Shanghai for Science and Technology Shanghai China The State Key Laboratory of Electronic Thin Films and Integrated Devices University of Electronic Science and Technology of China Chengdu China Group Department of Informatics Universität Hamburg Hamburg Germany The Institute of Machine Intelligence University of Shanghai for Science and Technology Shanghai China The State Key Lab. for Novel Software Technology Nanjing University Nanjing China The Engineering Research Center of Wideband Wireless Communication Technology Ministry of Education Nanjing University of Posts and Telecommunications Nanjing China The Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore The Department of Automation Tsinghua University Beijing China

In the classic setting of unsupervised domain adaptation (UDA), the labeled source data are available in the training phase. However, in many real-world scenarios, owing to some reasons such as privacy protection and information security, the source data is inaccessible, and only a model trained on the source domain is available. This paper proposes a novel deep clustering method for this challenging task. Aiming at the dynamical clustering at feature-level, we introduce extra constraints hidden in the geometric structure between data to assist the process. Concretely, we propose a geometry-based constraint, named semantic consistency on the nearest neighborhood (SCNNH), and use it to encourage robust clustering. To reach this goal, we construct the nearest neighborhood for every target data and take it as the fundamental clustering unit by building our objective on the geometry. Also, we develop a more SCNNH-compliant structure with an additional semantic credibility constraint, named semantic hyper-nearest neighborhood (SHNNH). After that, we extend our method to this new geometry. Extensive experiments on three challenging UDA datasets indicate that our method achieves state-of-the-art results. The proposed method has significant improvement on all datasets (as we adopt SHNNH, the average accuracy increases by over 3.0% on the large-scaled dataset). Code is available at https://***/tntek/N2DCX. Copyright © 2021, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Towards fine-grained compiler identification with neural modeling 32

Towards fine-grained compiler identification with neural mod...

引用

32nd International Conference on software Engineering and Knowledge Engineering, SEKE 2020

作者： Xie, Borun Tian, Zhenzhou Gao, Cong Chen, Lingwei School of Computer Science and Technology Xi'an University of Posts and Telecommunications Xi'an China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an China College of Information Sciences and Technology Pennsylvania State University PA United States

ISBN: (纸本)1891706500

Different compilers and optimization levels can be used to compile the source code. Revealed in reverse from the produced binaries, these compiler details facilitate essential binary analysis tasks, such as malware forensics and binary code similarity analysis. Most existing approaches adopt a signature matching based or machine learning based strategy to identify the compiler details, showing limits in either the detection accuracy or granularity. In this work, we propose NeuralCI (Neural modeling-based Compiler Identification) to perform the identification of compiler family and optimization level on individual functions. The basic idea is to formulate sophisticated neural networks to process abstracted instruction sequences generated using a lightweight function abstraction strategy. To evaluate the performance of NeuralCI, a large dataset consisting of 413,119 unique functions collected from real-world projects is constructed. The experiments show that NeuralCI achieves over 99% and 90% accuracy in identifying the compiler family and optimization level respectively, which outperforms most state-of-the-art function level compiler identification methods. Also, we explore for the first time the possibility of conducting compiler identification on binary code snippets rather than complete functions, where NeuralCI still achieves 96% accuracy, indicating its ability of capturing subtle yet significant features. © 2020 Knowledge Systems Institute Graduate School. All rights reserved.

关键词： Binary codes

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：