Find research datasets worth reusing

opencc-by-4.0Nov 2024View details →

Novel Libraries in Stack Overflow Posts

<p># Summary</p> <p>We present datasets detailing the appearance of novel libraries and library pairs in Stack Overflow posts in 12 languages between 2008 and 2023.</p> <div> <div># Disclaimer</div> <br> <div>Pair of libraries are displayed in the canonical format of <lib_a>|<lib_b> where lib_a precedes lib_b in alphabetical ordering.</div> <br> <div>Some of the examples are truncated for better readability.</div> <br> <div>GitHub source of the project: https://github.com/MeszarosGabor/SO_Post_Analyzer</div> <br> <div># Descriptions</div> <div>## `<language>`/all_`<language>`_so_posts.jsonl</div> <br> <div>JSONL file that contains the raw extracted Stack Overflow fields. Within a single JSON object:</div> <div>key: post_id,</div> <div>values:</div> <div>- post_type: 1 for question and 2 for answer</div> <div>- accepted_answer_id</div> <div>- date_posted</div> <div>- score</div> <div>- view_count</div> <div>- code_snippets</div> <div>- post_length</div> <div>- poster_id</div> <div>- last_actiivity</div> <div>- tags</div> <div>- number of comments</div> <div>- number of answers</div> <div>- parent id</div> <br> <div>Example:</div> <div>```</div> <div>{"72": ["1", "", "2008-08-01T13:38:27.133", "48", "2148", "<p>I want to format my existing comments as 'RDoc comments' so they can be viewed using <code>ri</code>.</p>\n\n<p>What are some recommended resources for starting out using RDoc?</p>\n", "25", "2016-12-30T06:56:18.310", "<ruby><rdoc>", "1", "2", ""]}</div> <div>```</div> <br> <div>## `<language>`/`<language>`_all_libs_dates.json</div> <br> <div>JSON file that lists the dates (with multiplicity, one for every post) when an individual library was mentioned in a post.</div> <br> <div>Example:</div> <div>```</div> <div>'FileUtils': ['2011-06-09',</div> <div>'2011-07-01',</div> <div>'2011-11-20',</div> <div>'2011-11-20',</div> <div>...</div> <div>'2013-09-04',</div> <div>'2020-05-08',</div> <div>'2021-02-25']</div> <div>```</div> <br> <div>## `<language>`/`<language>`_all_pairs_dates.json</div> <br> <div>JSON file that lists the dates (with multiplicity, one for every post) when a pair of libraries was mentioned in a post.</div> <br> <div>Example:</div> <div>```</div> <div>'mongo_mapper|sinatra': ['2010-09-12',</div> <div>'2011-12-30',</div> <div>'2012-02-23',</div> <div>'2012-09-04'],</div> <div>```</div> <br> <div>## `<language>`/`<language>`_libs_count.json</div> <br> <div>JSON file that lists the occurrence count of the individual libraries.</div> <br> <div>Example:</div> <div>```</div> <div>{</div> <div>'cairo': 4,</div> <div>'pango': 2,</div> <div>'radix': 1,</div> <div>}</div> <div>```</div> <br> <div>## `<language>`/`<language>`_pairs_count.json</div> <br><br> <div>JSON file that lists the co-occurrence count of the pairs of libraries.</div> <br> <div>Example:</div> <div>```</div> <div>'mongo_mapper|sinatra': 4,</div> <div>'fileutils|getoptlong': 1,</div> <div>'redis|rubygems': 24,</div> <div>```</div> <br> <div>## `<language>`/`<language>`_libs_first_dates.json</div> <br> <div>JSON file that lists the dates of the first appearances of individual libraries alongside the post id and poster id.</div> <br> <div>Example:</div> <div>```</div> <div>{</div> <div>'cairo': {'id': '6242589', 'poster_id': '784674', 'date': '2011-06-05'},</div> <div>}</div> <div>```</div> <br> <div>## `<language>`/`<language>`_pairs_first_dates.json</div> <div>JSON file that lists the dates of the first co-appearances of pairs libraries alongside the post id and poster id.</div> <br> <div>Example:</div> <div>```</div> <div>'rubygems|server': {'id': '3748309',</div> <div>'poster_id': '262808',</div> <div>'date': '2010-09-20'</div> <div>```</div> <br> <div>## `<language>`/`<language>`_`<language>`_code_count_list.json</div> <br> <div>JSON file that contains a single list of library counts in the posts (in chronological order) that contain *at least one* library import.</div> <br> <div>## `<language>`/`<language>`_daily_post_stats.json</div> <div>JSON file that counts the number of posts on a given day, listed chronologically, containing dates *with at least one post*. Dictionary of key=date value=count(int) pairs.</div> <br> <div>Example:</div> <div>```{...</div> <div>'2011-09-03': 6,</div> <div>'2011-09-04': 3,</div> <div>'2011-09-05': 10,</div> <div>'2011-09-06': 5,</div> <div>'2011-09-07': 15,</div> <div>...}</div> <div>```</div> <br> <div>## `<language>`/`<language>`_`<langugae>`_post_stats.json</div> <br> <div>JSON file that lists the individual post metadata (sorted by post date).</div> <div>Fields:</div> <div>- post id,</div> <div>- post type,</div> <div>- list of imports</div> <div>- post date</div> <div>- poster id</div> <div>- score</div> <br> <div>Example:</div> <br> <div>```</div> <div>{'id': '1892176',</div> <div>'post_type': '1',</div> <div>'imports': ['mechanize', 'rubygems'],</div> <div>'date': '2009-12-12T03:31:43.823',</div> <div>'poster_id': '124685',</div> <div>'score': '5'},</div> <div>```</div> <br> <div>## `<language>`/`<language>`_time_based_new.jsonl</div> <br> <div>JSONL file that contains JSON objects (in chronological order) detailing post metadata.</div> <div>Fields:</div> <div>- post id,</div> <div>- post date</div> <div>- poster id (user id)</div> <div>- post type,</div> <div>- list of imports</div> <div>- list of novel libraries in post</div> <div>- list of novel pairs in post</div> <br> <div>Example:</div> <div>```</div> <div>{'post_id': '3543',</div> <div>'post_date': '2008-08-06T15:24:00.787',</div> <div>'user_id': '399',</div> <div>'post_type': '2',</div> <div>'imports': ['metric_fetcher', 'rake'],</div> <div>'new_libs': ['metric_fetcher', 'rake'],</div> <div>'new_pairs': ['metric_fetcher|rake']}</div> <div>```</div> <br> <div>## `<language>`/`<language>`_user_to_posts.json</div> <br> <div>JSON file that lists the post ids corresponding to a given user id. Keyed by user ids, values are list of post ids.</div> <br> <div>Example:</div> <div>```</div> <div>'303675': ['2941479'],</div> <div>'348325': ['2945141', '2956990', '2968924', '3832703'],</div> <div>'325477': ['2945228'],</div> <div>'27196': ['2949100', '3177217'],</div> <div>```</div> </div>

opencc-by-4.0Mar 2022View details →

Perceptions on the utility of community question and answer websites like Stack Overflow to software developers (Replication package)

<p>Interview Questions on the perception of the utility of CQAs like Stack Overflow to software developers. In this study, we focused on the questions highlighted in yellow.</p>

opencc-by-sa-4.0Jan 2018View details →

Usage and Attribution of Stack Overflow Code Snippets in GitHub Projects — Supplementary Material

<p><em>Background:</em> Stack Overflow (SO) is the largest Q&A website for software developers, providing a huge amount of copyable code snippets. Using those snippets raises various maintenance and legal issues. SO’s license (CC BY-SA 3.0) requires attribution, i.e., referencing the original question or answer, and requires derived work to adopt a compatible license. While there is a heated debate on SO’s license model for code snippets and the required attribution, little is known about the extent to which snippets are copied from SO without proper attribution.</p> <p><em>Aim:</em> Our main goal was to analyze how often code from SO posts is used in public GitHub projects, but not attributed as required by the license. Further, we wanted to investigate if developers are aware of SO’s license and its implications, and to what degree they adhere to the attribution requirements defined in SO’s terms of service.</p> <p><em>Method:</em> We present results of a large-scale empirical study analyzing the usage and attribution of non-trivial Java code snippets from SO answers in public GitHub projects. We followed three different approaches to triangulate an estimate for the ratio of unattributed usages and conducted two online surveys with software developers to complement our results.</p> <p><em>Results:</em> For the different sets of projects that we analyzed, the amount of projects containing files with a reference to SO varied between 3.3% and 11.9%. We found that at most 1.8% of all analyzed repositories containing code from SO used the code in a way compatible with CC BY-SA 3.0. Moreover, we estimate that at most a quarter of the copied code snippets from SO are attributed as required, i.e., using a link in a source code comment. About half of the surveyed developers admitted copying code from SO without attribution. Furthermore, about two thirds of them were not aware of the license of SO code snippets and its implications.</p>

opencc-by-sa-4.0Mar 2018View details →

SOTorrent: Reconstructing and Analyzing the Evolution of Stack Overflow Posts — Supplementary Material

<p>Stack Overflow is the most popular question-and-answer website for software developers, providing a large amount of code snippets and free-form text on a wide variety of topics. Like other software artifacts, questions and answers on Stack Overflow evolve over time, for example when bugs in code snippets are fixed, code is updated to work with a more recent library version, or text surrounding a code snippet is edited for clarity. To be able to analyze how content on Stack Overflow evolves, we built <em>SOTorrent</em>, an open dataset based on the official Stack Exchange data dump. <em>SOTorrent </em>provides access to the version history of Stack Overflow content at the level of whole posts and individual text or code blocks.</p> <p>This dataset has been retrieved from <em>SOTorrent </em>using the following scripts:</p> <p><a href="https://doi.org/10.5281/zenodo.1201679">https://doi.org/10.5281/zenodo.1201679</a></p> <p>For the MSR 2018 paper about SOTorrent, we used the following scripts to analyze the data:</p> <p><a href="https://doi.org/10.5281/zenodo.1201706">https://doi.org/10.5281/zenodo.1201706</a></p> <p>The files <em>sample_before_10.ods</em> and <em>sample_after_10.ods</em> contain our qualitative analysis of 50 comments that were made up to 10 minutes before/after an edit.</p>

opencc-by-4.0Aug 2022View details →

Diverse Title Generation for Stack Overflow Posts with Multiple Sampling Enhanced Transformer

<p>Dataset for our paper "Diverse Title Generation for Stack Overflow Posts with Multiple Sampling Enhanced Transformer"</p> <p>(<a href="https://github.com/zfj1998/M3NSCT5">zfj1998/M3NSCT5: the code base for our paper "Diverse Title Generation for Stack Overflow Posts with Multiple Sampling Enhanced Transformer" (github.com)</a>)</p> <p>Including three files representing the train/val/test datasets. Each file contains all the collected data covering eight programming languages.</p>

opencc-by-4.0Dec 2023View details →

threads-stack-overflow

<h3>Overview</h3> <p>This is a temporal higher-order network dataset, which here means a sequence of timestamped hyperedges where each hyperedge is a set of nodes. In this dataset, nodes are users on <a href="https://stackoverflow.com/">stackoverflow.com</a>, and a hyperedge comes from users participating in a thread that lasts for at most 24 hours. The timestamps are the time of the post, but normalized so that the earliest post starts at 0. </p> <h3><strong>Source of original data</strong></h3> <p>Source: <a href="https://www.cs.cornell.edu/~arb/data/threads-stack-overflow/">threads-stack-overflow dataset</a></p> <h3><strong>References</strong></h3> <p>If you use this data, please cite the following paper:</p> <ul> <li><a href="https://doi.org/10.1073/pnas.1800683115">Simplicial closure and higher-order link prediction</a>. Austin R. Benson, Rediet Abebe, Michael T. Schaub, Ali Jadbabaie, and Jon Kleinberg. Proceedings of the National Academy of Sciences (PNAS), 2018.</li> </ul>

opencc-by-4.0Apr 2022View details →

Research Artefact: An Empirical Study of React-Library Related Issues via Stack Overflow

<p>This research artifact accompanies the paper titled "An Empirical Study of React-Library Related Issues via Stack Overflow." It is a comprehensive repository that includes the collected dataset containing 447,542 React-related Stack Overflow question posts, as well as 384 representative samples obtained randomly. The primary objective of this artifact is to facilitate the replication of our dataset for researchers and allow them to utilize it for further investigations and research purposes.</p>

opencc-by-4.0May 2018View details →

Data for "Is Stack Overflow in Portuguese attractive for Brazilian Users?"

<p>Data for Botto-Tobar et al. Is Stack Overflow in Portuguese attractive for Brazilian Users?. ICGSE 2018.</p> <p>This data was built based on data dump from Stack Exchange (https://stackexchange.com) website. It contains two separate databases (Stack Overflow in English and Stack Overflow in Portuguese):</p> <ul> <li>Users</li> <li>Posts, decomposed by Answers and Questions</li> <li>Tags</li> <li>PostTags</li> <li>GenderUser</li> <li>UserLocation</li> </ul> <p>For more information, please visit http://www.win.tue.nl/~mbottoto/files/papers/conference_papers/sopt_icgse2018.pdf or write to <em>m.a.botto.tobar@tue.nl</em></p> <p> </p>

opencc-by-4.0Feb 2019View details →

Pylint Results for Python Code Snippets on Stack Overflow

<p>This dataset contains valid pylint results for all Stack Overflow code snippets from SOTorrent that meet the following criteria</p> <ul> <li>Tagged with 'python'</li> <li>6 lines and above</li> <li>Contains basic python syntax (i.e. 'print', 'import', '(', '=')</li> <li>Produces a result when processed by Pylint</li> </ul>

opencc-by-4.0Aug 2019View details →

Replication package for the paper: "Technical Debt's State of Practice on Stack Overflow: a Preliminary Study"

<p>This is the replication package for the paper "Technical Debt’s State of Practice on Stack Overflow: a Preliminary Study", published (in Portuguese) in the preliminary results track of SBQS, the Brazilian Symposium on Software Quality.</p> <p> </p> <p>We provide the data for all steps of our methodology and final analysis. Each file is numbered, indicating the order in which they were produced in our study. </p>

opencc-by-4.0Jul 2021View details →

Research Artefact: What network simulator questions do users ask? a large-scale study of stack overflow posts

<p><strong>Research Artefact: What network simulator questions do users ask? a large-scale study of stack overflow posts</strong></p> <p>This is a research artefact for the paper: <strong>What network simulator questions do users ask? a large-scale study of stack overflow posts</strong>. This artefact is a repository consisting of the collected dataset including 2,322 network-simulator-related Stack Overflow questions. This artefact aims to enable researchers to replicate our dataset of the paper and reuse the dataset for further research.</p>

opencc-by-4.0Aug 2021View details →

How Do I Refactor This? An Empirical Study on Refactoring Trends and Topics in Stack Overflow

<p>This is the dataset that accompanies the study: "<strong>How Do I Refactor This? An Empirical Study on Refactoring Trends and Topics in Stack Overflow</strong>." This study has been accepted for publication in the software engineering journal Empirical Software Engineering.</p> <p> </p> <p><strong><em>Following is the abstract of the study:</em></strong></p> <p>An essential part of software maintenance and evolution, refactoring is performed by developers, regardless of technology or domain, to improve the internal quality of the system, and reduce its technical debt. However, choosing the appropriate refactoring strategy is not always straightforward, resulting in developers seeking assistance. Although research in refactoring is well-established, with several studies altering between the detection of refactoring opportunities and the recommendation of appropriate code changes, little is known about their adoption in practice. Analyzing the perception of developers is critical to understand better what developers consider to be problematic in their code and how they handle it. Additionally, there is a need for bridging the gap between refactoring, as research, and its adoption in practice, by extracting common refactoring intents that are more suitable for what developers face in reality. In this study, we analyze refactoring discussions on Stack Overflow through a series of quantitative and qualitative experiments. Our results show that Stack Overflow is utilized by a diverse set of developers for refactoring assistance for a variety of technologies. Our observations show five areas that developers typically require help with refactoring-- Code Optimization, Tools and IDEs, Architecture and Design Patterns, Unit Testing, and Database. We envision our findings better bridge the support between traditional (or academic) aspects of refactoring and their real-world applicability, including better tool support.</p> <p><strong>Manuscript DOI:</strong> <a href="https://doi.org/10.1007/s10664-021-10045-x">https://doi.org/10.1007/s10664-021-10045-x</a> </p>

opencc-by-4.0Apr 2020View details →

Usuarios mejor valorados en Stack overflow

<p>Dataset que recopila la información más importante a para la evaluación de un usuario de la plataforma Stack overflow. En este dataset se presenta los datos más relevantes para cada uno de los usuarios pertenecientes al top 50 en reputación de este año en curso 2020.</p>

opencc-by-4.0Apr 2017View details →

Replication package for How the R Community Creates and Curates Knowledge: An Extended Study of Stack Overflow and Mailing Lists

<p>This dataset was used in the paper: "How the R Community Creates and Curates Knowledge: An Extended Study of Stack Overflow and Mailing Lists", Journal of Empirical Software Engineering, to appear.</p>

opencc-by-4.0Nov 2023View details →

Replication Package: Vulnerably (Mis)Configured? Exploring 10 Years of Developers' Q&As on Stack Overflow

<p><strong>Welcome to the public repository for the additional content of the paper "Vulnerably (Mis)Configured? Exploring 10 Years of Developers' Q&As on Stack Overflow", accepted at the International Working Conference on Variability Modelling of Software-Intensive Systems (VAMOS) 2024.</strong></p><p>This repository provides additional information to the conducted exploratory study on configuration-related vulnerabilities, including the following files:</p><ul><li>README.txt</li><li>LICENSE.txt</li><li>DATASET_CONFIG_VULN_SO.csv: sheet containing data of 651 StackOverflow posts, including additional classifications based on manual analyses and automatic topic modeling</li></ul><p><strong>Instructions for using the dataset</strong></p><ol><li>Download and open the dataset (platform-independent CSV file).</li><li>The dataset includes 16 columns (A – P):<br>- Columns A – J: Original data fetched from the BigQuery Stack Overflow dataset (<i>Question_ID, Year_Asked, Question_Title, Question_Body, Question_Tags, View_Count, Question_Rating, Favorite_Count, Status, Answer_Count</i>)<br>- Columns K – N: Manually extracted data from the Stack Overflow posts (<i>System, Configuration Context, Security Context, Topic</i>)<br>- Column O: Data based on the automated topic modeling (<i>Configuration Topic</i>)<br>- Column P: Additional data extracted from the Stack Overflow posts without further classifications (<i>Additional Comments</i>)</li></ol><p><strong>Requirements</strong></p><ul><li>No requirements</li></ul><p><strong>Further information</strong></p><ol><li>The dataset is based on a search string (SQL query; August 1, 2023) applied on the Google BigQuery Stack Overflow dataset:<i> </i><br><i>("secur*") AND ("vulnerabilit*" OR "weakness*" OR "breach*" OR "exposure*" OR "CVE*" OR "CWE*") AND ("config*")</i></li><li>Originally, the dataset included 1,235 post which were limited by the first and second authors to 651 posts (34 deleted posts, 550 posts out of scope) using the following selection criteria: <br>- The post has been created in the last decade (2013-2022).<br>- The post is still available on the Stack Overflow website.<br>- The post is directly connected to a vulnerability-related issue in the context of configuring.</li><li>Topic modeling algorithm used: Latent Dirichlet Allocation (LDA)<br>- Settings: 200 iterations (coherence value = 0.6 for k = 7 to 11), α = k, β = 0.01</li></ol>

opencc-by-4.0Aug 2023View details →

Harmonising Contributions: Exploring Diversity in Software Engineering through CQA Mining on Stack Overflow

<p>Community question-and-answering platforms dedicated to software engineering, such as Stack Overflow, have assumed indispensable roles in fostering a thriving global knowledge ecosystem. As these platforms suffer from diversity-related issues, investigating the underlying reasons behind such challenges becomes imperative to devise potential intervention strategies.</p> <p>The proposed study highlights Stack Overflow users’ contribution profiles, both in isolation and relative to various diversity metrics, including GDP and access to electricity. Finally, the study explores whether these contribution profiles extend to the city and state levels.</p> <p>This replication package complements our study, prompting future scholars to further examine our research process or conduct follow up analyses.</p>

opencc-by-4.0Apr 2024View details →

A Cross-Continental Analysis of How Regional Cues Shape Developers' Stack Overflow Contributions

<p>Stack Overflow provides a wide range of knowledge for the software development community. Despite the importance of these platforms, several studies have shown that digital information tends to cluster geographically, which limits knowledge access that is otherwise necessary for innovation.</p> <p>The proposed study highlights the dynamics of users from different geographical backgrounds within Stack Overflow, which entails intra-country interactions, predominant topics of discouse, as well as their communication patterns. Finally, the study highlights that regional behavioural variations stem beyond cultural factors, encompassing technological advancement, entrepreneurial ventures, and workforce composition. </p> <p>This replication package is provided for those interested in further examining our research methodology.</p>

opencc-by-4.0Aug 2024View details →

The Influence of Code Comments on the Perceived Helpfulness of Stack Overflow Posts (Supplementary Material)

<p>Question-and-answer platforms such as Stack Overflow have become an important way for software developers to share and retrieve knowledge. However, reusing poorly understood code can lead to serious problems, such as bugs or security vulnerabilities. To better understand the role of code comments in the perceived helpfulness of answers on Stack Overflow, we conducted an online experiment simulating a Stack Overflow environment (n=91). The results indicate that block comments in particular were perceived as significantly more helpful than uncommented source code, especially by novices. Novices also found code snippets with block comments more helpful than those with inline comments. Interestingly, other surface features, such as the position of an answer and its usefulness score, were considered less important. The content of Stack Overflow has been a major source for training large language models. Although AI-based coding assistants such as GitHub Copilot, which are based on these models, might change the way Stack Overflow is used, our findings have implications beyond this specific platform. First, the findings of this study may help improve the ongoing relevance of community-driven platforms like Stack Overflow that provide human advice and explanations of code solutions to complement AI-based support for programmers. Second, since chat-based tools can be prompted to generate code in different ways, knowing which properties influence the perceived helpfulness of code snippets is beneficial.</p>