检索结果-南通市图书馆

State and tendency: an empirical study of deep learning question&answer topics on stack overflow

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2021年第11期64卷 131-153页

作者： Henghui ZHAO Yanhui LI Fanwei LIU Xiaoyuan XIE Lin CHEN State Key Laboratory for Novel Software Technology Nanjing University School of Computer Science Wuhan University

Deep learning has developed rapidly in recent years, attracting the attention of numerous researchers. Since a wide range of topics are covered in this field, we are wondering what topics researchers have concerned about. However, after investigation, we find that very few researchers have paid attention to this demand. In this paper, we conduct a large-scale study to analyze the questions faced by deep learning developers. We use stack overflow, one of the largest question&answer sites, as our data source, and extract32969 posts about deep learning as our studied dataset. After filtering, augmenting and pre-processing the post datasets from stack overflow, we use the latent Dirichlet allocation(LDA) topic model to summarize30 topics based on their text content. In addition, we measure the difficulty and popularity of each topic,compare the different issues faced by different deep learning frameworks, and analyze the development trend of each topic. Our main results are as follows:(1) developers ask a broad spectrum of questions about deep-learning, ranging from Data Shape to Object Detection;(2) Gradient Propagation is the most popular among all the topics and(3) Object Detection is the most difficult;(4) issues of Package Installation, Code Understanding and Method Introduction are common in the current different deep learning frameworks;(5) there are three trends in these topics, e.g., a significant rising trend is found in the number of discussion on Data Shape. Finally, based on our research findings, we make some targeted and valuable suggestions for developers, researchers, educators, and framework providers of deep learning.

关键词： topic model deep learning stack overflow question&answer

IEA: an answerer recommendation approach on stack overflow

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2019年第11期62卷 51-69页

作者： Liting WANG Li ZHANG Jing JIANG State Key Laboratory of Software Development Environment Beihang University

stack overflow is a web-based service where users can seek information by asking questions and share knowledge by providing answers about software development. Ideally, new questions are assigned to experts and answered within a short time after their submissions. However, the number of new questions is very large on stack overflow, answerers are not easy to find suitable questions timely. Therefore, an answerer recommendation approach is required to assign appropriate questions to answerers. In this paper, we make an empirical study about developers' activities. Empirical results show that 66.24% of users have more than30% of comment activities. Furthermore, active users in the previous day are likely to be active in the next day. In this paper, we propose an approach IEA which combines user topical interest, topical expertise and activeness to recommend answerers for new questions. We first model user topical interest and expertise based on historical questions and answers. We also build a calculation method of users' activeness based on historical questions, answers, and comments. We evaluate the performance of IEA on 3428 users containing41950 questions, 64894 answers, and 96960 comments. In comparison with the state-of-the-art approaches of TEM, TTEA and TTEA-ACT, IEA improves n DCG by 2.48%, 3.45% and 3.79%, and improves Pearson rank correlation coefficient by 236.20%, 84.91% and 224.12%, and improves Kendall rank correlation coefficient by 424.18%, 1845.30% and 772.60%.

关键词： answerer recommendation activeness comments topical interest topical expertise stack overflow

Multi-Factor Duplicate Question Detection in stack overflow

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2015年第5期30卷 981-997页

作者：张芸 David Lo 夏鑫孙建伶 College of Computer Science and Technology Zhejiang University Hangzhou 310027 China School of Information Systems Singapore Management University Singapore Singapore

stack overflow is a popular on-line question and answer site for software developers to share their experience and expertise. Among the numerous questions posted in stack overflow, two or more of them may express the same point and thus are duplicates of one another. Duplicate questions make stack overflow site maintenance harder, waste resources that could have been used to answer other questions, and cause developers to unnecessarily wait for answers that are already available. To reduce the problem of duplicate questions, stack overflow allows questions to be manually marked as duplicates of others. Since there are thousands of questions submitted to stack overflow every day, manually identifying duplicate questions is a difficult work. Thus, there is a need for an automated approach that can help in detecting these duplicate questions. To address the above-mentioned need, in this paper, we propose an automated approach named DuPPREDICTOR that takes a new question as input and detects potential duplicates of this question by considering multiple factors. DuPPREDICTOR extracts the title and description of a question and also tags that are attached to the question. These pieces of information （title, description, and a few tags） are mandatory information that a user needs to input when posting a question. DuPPREDICTOR then computes the latent topics of each question by using a topic model. Next, for each pair of questions, it computes four similarity scores by comparing their titles, descriptions, latent topics, and tags. These four similarity scores are finally combined together to result in a new similarity score that comprehensively considers the multiple factors. To examine the benefit of DuPPREDICTOR, we perform an experiment on a stack overflow dataset which contains a total of more than two million questions. The result shows that DuPPREDICTOR can achieve a recali-rate@20 score of 63.8%. We compare our approach with the standard search engine of stack overflow, and DuPPREDICTOR improves its recall-rate@10

关键词： software information site duplicate question stack overflow DupPredictor

What Security Questions Do Developers Ask？ A Large-Scale Study of stack overflow Posts

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2016年第5期31卷 910-924页

作者： Xin-Li Yang David Lo Xin Xia Zhi-Yuan Wan Jian-Ling Sun College of Computer Science and Technology Zhejiang University Hangzhou 310027 China School of Information Systems Singapore Management University Singapore Singapore

Security has always been a popular and critical topic. With the rapid development of information technology, it is always attracting people＇s attention. However, since security has a long history, it covers a wide range of topics which change a lot, from classic cryptography to recently popular mobile security. There is a need to investigate security-related topics and trends, which can be a guide for security researchers, security educators and security practitioners. To address the above-mentioned need, in this paper, we conduct a large-scale study on security-related questions on stack overflow. stack overflow is a popular on-line question and answer site for software developers to communicate, collaborate, and share information with one another. There are many different topics among the numerous questions posted on stack overflow and security-related questions occupy a large proportion and have an important and significant position. We first use two heuristics to extract from the dataset the questions that are related to security based on the tags of the posts. And then we use an advanced topic model, Latent Diriehlet Allocation （LDA） tuned using Genetic Algorithm （GA）, to cluster different security-related questions based on their texts. After obtaining the different topics of security-related questions, we use their metadata to make various analyses. We summarize all the topics into five main categories, and investigate the popularity and difficulty of different topics as well. Based on the results of our study, we conclude several implications for researchers, educators and practitioners.

关键词： security stack overflow empirical study topic model

Geek Talents:Who are the Top Experts on GitHub and stack overflow?

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2019年第8期61卷 465-479页

作者： Yijun Tian Waii Ng Jialiang Cao Suzanne McIntosh New York University Courant Institute of Mathematical SciencesNew York10012USA

In the field of Computer Science,software developers need to use a wide array of social collaborative platforms for learning and cooperating.The most popular ones are GitHub and stack overflow.Existing platforms only support search queries to extract relevant repository information from GitHub,or questions and answers from stack overflow.This ignores the valuable coder-related part-who are the top experts(geek talents)in a specific area?This information is important to companies,open source projects,and to those who want to learn from an expert role model.Thus,how to find the right developers is quite a crucial yet challenging problem.Most of the current works mainly focus on recommending experts in a particular software engineering task and ignore the relationship between developers within different projects.In this paper,we propose a novel technique that automatically identifies geek talents from GitHub,stack overflow,and across both communities.The results show that our work performs well at recommending proper developers in diverse areas.

关键词： Developer recommendation collaborative filtering stack overflow GitHub

一种结合代码片段和混合主题模型的软件数据聚类方法

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

计算机科学 2024年第6期51卷 44-51页

作者：魏林林沈国华黄志球蔡梦男郭菲菲南京航空航天大学计算机科学与技术学院南京211106 南京航空航天大学高安全系统的软件开发与验证技术工业和信息化部重点实验室南京211106 软件新技术与产业化协同创新中心南京210093

使用主题模型进行文档聚类是众多文本挖掘任务中一种常见的做法。许多研究针对软件问答网站的数据,利用主题模型进行聚类来分析不同领域在社区的发展情况。然而,这些软件相关数据往往包含代码片段且文本长度分布不均,使用传统单一的主... 详细信息

使用主题模型进行文档聚类是众多文本挖掘任务中一种常见的做法。许多研究针对软件问答网站的数据,利用主题模型进行聚类来分析不同领域在社区的发展情况。然而,这些软件相关数据往往包含代码片段且文本长度分布不均,使用传统单一的主题模型对文本数据建模,易得到不稳定的聚类结果。文中提出了一种结合代码片段和混合主题模型的聚类方法,并使用stack overflow作为数据源,构造了在该平台上被提问数量排名前60的Python第三方库数据集,经过建模,该数据集最终划分为以下6个不同的领域:网络安全、数据分析、人工智能、文本处理、软件开发和系统终端。实验结果表明,在自动评估和人工评估的指标上,使用代码片段结合文本进行主题建模,在聚类结果划分的质量上表现良好,而联合多个模型进行实验,一定程度上提高了聚类结果的稳定性和准确性。

关键词：代码片段主题模型 stack overflow Python 聚类

维普期刊数据库

基于stack overflow的并发漏洞主题分析与系统实现

在线全文

学校读者我要写书评

暂无评论

基于Stack Overflow的并发漏洞主题分析与系统实现

作者：孟醒扬州大学

学位级别：硕士

随着信息技术的不断发展,计算机软件的安全问题受到了广泛的关注,其中漏洞是造成软件安全威胁的主要原因之一。如今随着多核操作系统的流行,多线程程序在执行过程中由于对共享资源的竞争而导致并发漏洞,一旦并发漏洞被利用,可能会导致... 详细信息

随着信息技术的不断发展,计算机软件的安全问题受到了广泛的关注,其中漏洞是造成软件安全威胁的主要原因之一。如今随着多核操作系统的流行,多线程程序在执行过程中由于对共享资源的竞争而导致并发漏洞,一旦并发漏洞被利用,可能会导致如程序崩溃、拒绝服务攻击、数据损坏等后果,严重影响到系统的可用性、完整性和安全性,因此并发漏洞的研究对解决网络安全问题具有重要意义。stack overflow作为流行的技术问答社区,包含了大量与并发漏洞相关的问答帖子,如何从大量文本中挖掘用户关注的并发漏洞相关的需求成为了研究重点。随着神经网络的发展,神经主题模型在主题挖掘中开始应用,但现有的神经主题模型存在未能充分利用语料库中的外部知识,无法充分利用文档间的关系推断主题等问题。为了帮助用户更好了解并发漏洞帖子内容,本文以stack overflow中与并发漏洞领域相关的问题帖作为研究对象,对当前神经主题模型中存在的不足进行改进,力求能够更加准确的挖掘并发漏洞主题和完成主题分析任务。针对以上问题,本文的主要工作如下:1.考虑到基于双向对抗训练的高斯神经主题模型编码器中的Glove静态词向量导致的单词语义表示质量不佳的问题,本论文中采用结合动态上下文语义学习的ALBERT预训练模型对文档向量表征进行改进,进一步提升语义表征质量,并引入CycleGAN循环一致对抗损失更好地引导编码器和生成器网络进行对抗训练,通过所构建的数据集进行对比实验,验证了本文中提出模型的有效性,并得到更具解释性的主题词生成表示。2.传统的主题模型在训练过程中仅利用了单个文档内词语的相互关系,未能充分利用与该文档主题相关的其他文档和词语信息。本文提出基于图注意力网络的主题建模方法GATTM,通过把所有文档和词语看作结点的方式进行连接,将并发漏洞语料库中文档和词语表示为文档关系图,并引入GATv2网络对注意力功能进行修正,进一步增强了节点的特征向量表示,对文档的邻接结点进行信息聚合来识别当前文档的主题分布。通过所构建的数据集进行对比实验,结果发现GATTM在主题建模性能上优于其它模型。3.本研究中构建了一个基于stack overflow的并发漏洞数据集,在经过数据获取,设置初始标签,标签扩充,数据预处理等过程后,共得到47683条并发漏洞帖子,作为最终的实验数据集,基于本文提出的主题模型对并发漏洞相关帖子进行统计分析和主题分析,讨论各个主题下的流行度和难度排名,并针对主题流行度和难度结果进行主题分析,希望发现并发漏洞领域的热点和难点问题,为并发开发者和研究人员确定工作重点提供决策。4.基于本研究所提出的主题模型,设计和实现了一个并发漏洞领域主题分析系统,系统主要分为用户注册登录、数据集上传,主题建模、主题分析和用户下载管理等功能,通过上传并发漏洞数据集,帮助用户了解热点和难点问题。

关键词：并发漏洞 stack overflow 神经网络主题模型主题分析

基于注意力机制的stack overflow问题相关性预测模型研究

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

基于注意力机制的Stack Overflow问题相关性预测模型研究

作者：裴嘉言华南理工大学

学位级别：硕士

stack overflow是最流行的编程社区问答(PCQA)网站之一,近年来吸引了越来越多的用户。用户的增加使得网站上有大量的问题被提出,而这些问题之间可能存在相关性而未被识别。当用户在stack overflow提出或查询问题时,若能提供与该问题相... 详细信息

stack overflow是最流行的编程社区问答(PCQA)网站之一,近年来吸引了越来越多的用户。用户的增加使得网站上有大量的问题被提出,而这些问题之间可能存在相关性而未被识别。当用户在stack overflow提出或查询问题时,若能提供与该问题相关的问题有助于用户解决问题。此外,识别出重复问题能够更好地帮助网站进行维护,也减少了提出重复问题的用户的不必要等待时间。尽管已有大量研究方法致力于自动预测stack overflow上问题之间的相关性,但这些方法中,基于特征的方法未能充分利用问题中的语义信息,并且无法克服词汇间隙问题。而深度学习方法大多基于句子编码模型的结构进行构建,这类方法可能会丢失两个问题之间的语义交互信息。此外,由于问题内通常包含着较长的文本序列,该类模型也无法对文本进行合适的特征提取以及长距离依赖信息的有效捕获。针对上述问题,本文中提出了基于注意力的句子对交互模型(Attention-based Sentence Interaction Model)ASIM,用于自动预测stack overflow上问题之间的相关性。ASIM模型主要由(1)文本编码及特征提取框架和(2)交互信息提取与融合框架构成。在文本编码及特征提取框架中,本文通过预训练特定于软件工程领域的词嵌入,减少一般领域词嵌入在大量与软件工程无关的语料上训练带来的单词含义不准确问题,并基于自注意力机制对双向LSTM编码器的输出进行问题内交互信息的提取以及长距离依赖信息的捕获;在交互信息提取与融合框架中,本文采用软注意力机制对问题间交互信息进行提取,并通过与原始问题特征的融合,使得模型做出更稳定的匹配,进一步提升预测效果。实验结果表明,ASIM在stack overflow知识单元数据集上获得了超越其他模型的优秀性能,相较于此前效果最好的模型DOTBILSTM,在Micro-F1评估指标上有着6.9%的绝对性能提升,达到了81.91%的得分。此外,为验证模型的泛化性能,ASIM在软件工程领域的Ask Ubuntu重复问题检测数据集、一般领域中的Quora Question Pairs数据集、Sci Tail数据集以及Wiki QA数据集上进行了实验。ASIM同样保持了优越的性能,超越了众多现有方法,证明了该模型的泛化性以多领域预测能力,为模型的理论和设计提供了有力的支持。

关键词：自然语言处理 stack overflow 问题相关性预测注意力机制词嵌入

stack overflow的缺陷代码特征分析与相似缺陷检测

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

小型微型计算机系统 2021年第3期42卷 661-665页

作者：亢振兴赵逢禹刘亚上海理工大学光电信息与计算机工程学院上海200093

目前在软件代码缺陷审查以及缺陷预测中,研究人员对源代码进行分析研究却忽略了代码的缺陷信息.本文通过对缺陷信息进行分析,发现缺陷信息对于相似缺陷的检测有着重要的参考价值.基于这一思想,本文分析软件缺陷社区stack overflow中关... 详细信息

目前在软件代码缺陷审查以及缺陷预测中,研究人员对源代码进行分析研究却忽略了代码的缺陷信息.本文通过对缺陷信息进行分析,发现缺陷信息对于相似缺陷的检测有着重要的参考价值.基于这一思想,本文分析软件缺陷社区stack overflow中关于缺陷代码的信息,提出一种基于缺陷代码特征分析的相似缺陷检测方法.该方法首先对缺陷报告进行LDA主题分析并将缺陷报告分类到不同的主题(类别)中,统计得到高频缺陷类别;其次对于高频缺陷类别的缺陷代码提取特征;最后根据缺陷代码特征构建相似缺陷检测模型.为了验证相似缺陷检测模型的有效性,针对数据操作缺陷数据构建诊断模型并对该模型进行实证,实验结果表明该方法对检测其他代码中相似缺陷有较好的效果.

关键词： stack overflow LDA 缺陷代码特征特征相似度相似缺陷检测

维普期刊数据库