Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
72
datasets available to search
ShareScore release 0.9.0
Dataset results
72 results for “stack overflow”
StackPilot: Contrasting Code Snippets from Stack Overflow and GitHub Copilot
<p>Copy-paste programming via Stack Overflow and code generation via GitHub Copilot both define a query/prompt-based programming model. To enable systematic comparison of code copied from Stack Overflow and code generated by GitHub Copilot, we provide a dataset of 30,746 code snippets that Stack Overflow and GitHub Copilot produced in response to the same 2,636 queries/prompts.</p>
Stack Overflow's Hidden Nuances: How Does Zip Code Define User Contribution?
<p>Collective intelligence constitutes a foundational element within online community question-and-answering (CQA) platforms, such as Stack Overflow, being the source of most programming-related issues. Despite this relevance, concerns remain regarding issues surrounding user participation. Precedent research tends to focus on simple numerical measurements to analyse participation, which may sideline the inherent, subtler aspects.</p> <p>The proposed study aims to bridge this gap by operationalising 11 distinct metrics to represent user participation, behaviour, and community value across different regions of the USA. The study also conducts inductive content analysis to understand the impact of regional contextual factors on users' knowledge sharing patterns.</p> <p>This replication package is provided for those interested in further examining our research methodology.</p>
Stack Overflow's Hidden Nuances: How Does Zip Code Define User Contribution?
<p>--Collective intelligence constitutes a foundational element within online community question-and-answering (CQA) platforms, such as Stack Overflow, being the source of most programming-related issues. Despite this relevance, concerns remain regarding issues surrounding user participation. Precedent research tends to focus on simple numerical measurements to analyse participation, which may sideline the inherent, subtler aspects.</p> <p>The proposed study aims to bridge this gap by operationalising 11 distinct metrics to represent user participation, behaviour, and community value across different regions of the USA. The study also conducts inductive content analysis to understand the impact of regional contextual factors on users' knowledge sharing patterns.</p> <p>This replication package is provided for those interested in further examining our research methodology.</p>
Harmonising Contributions: Exploring Diversity in Software Engineering through CQA Mining on Stack Overflow – Replication Package
<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>
Does Location Influence Coding Practices? A Cross-Regional Study on Stack Overflow Code Quality
<p>Developers routinely integrate Stack Overflow code snippets into their codebases. However, the quality of snippets embedded in users’ answers remain elusive, and existing evaluations of code quality tend to be language or context-specific. Moreover, literature have found that contribution patterns vary depending on geographical locales, creating an unexplained rift between code quality, user location, and latent contextual regional factors. </p> <p>The proposed study evaluates the quality of SQL, JavaScript, Python, Ruby, and Java snippets across reliability, readability, performance, and security dimensions, benchmarking findings across states in the USA and investigating how different diversity indicators correlate against code quality violations. The study culminates in a series of inductive content analyses that qualitatively supplement prior quality dimensions.</p> <p>This replication package is provided for those interested in further examining our research methodology.</p>
Coding data of manuscript "How Do Developers Utilize Source Code from Stack Overflow?"
<p>This is the coding data for the manuscript "How Do Developers Utilize Source Code from Stack Overflow?".</p>
Supplementary material for "can you tell me if it smells? A study on how developers discuss code smells and anti-patterns in Stack Overflow"
<p>This dataset contains all data and results from the "Can you tell me if it smells? A study on how developers discuss code smells and anti-patterns in Stack Overflow" paper that was accepted at the 22nd International Conference on Evaluation andAssessment in Software Engineering (EASE), Christchurch, New Zealand.</p>
Contextual Documentation Referencing on Stack Overflow — Supplementary Material
<p>Supplementary material for our paper "Contextual Documentation Referencing on Stack Overflow".</p>
Dataset with manually validated version histories of Stack Overflow posts
<p>We used this dataset to evaluate different string similarity metrics for SOTorrent (http://sotorrent.org/). For the versions published 2018-11-01 and 2018-12-14, we double-checked and updated the ground truth files.</p> <p>The dataset has been created with this tool: https://github.com/sotorrent/posthistory-gt</p> <p>The dataset has been validated with this tool: https://github.com/sotorrent/posthistory-comparator-gt-cs</p> <p>The dataset has been used in this project: https://github.com/sotorrent/metric-evaluation</p> <p>The most recent version of the files can always be found here: https://github.com/sotorrent/metric-evaluation/tree/master/testdata/samples_comparison</p>
Does Location Influence Code Quality? Mining Stack Overflow Snippets Across the United States – Replication Package
<p>Developers routinely integrate Stack Overflow code snippets into their codebases. However, the quality of snippets embedded in users’ answers remain elusive, and existing evaluations of code quality tend to be language or context-specific. Moreover, literature have found that contribution patterns vary depending on geographical locales, creating an unexplained rift between code quality, user location, and latent contextual regional factors. </p> <p>The proposed study evaluates the quality of SQL, JavaScript, Python, Ruby, and Java snippets across reliability, readability, performance, and security dimensions, benchmarking findings across states in the USA and investigating how different diversity indicators correlate against code quality violations. The study culminates in a series of inductive content analyses that qualitatively supplement prior quality dimensions.</p> <p>This replication package is provided for those interested in further examining our research methodology.</p>
Stack Overflow
Open the record for dataset details and reuse information.
Artifacts for the FDL21 Paper: In-Vivo Stack Overflow Detection and Stack Size Estimation for Low-End Multithreaded Operating Systems using Virtual Prototypes
<p>Artifacts for the evaluation of the paper <em>In-Vivo Stack Overflow Detection and Stack Size Estimation for Low-End Multithreaded Operating Systems using Virtual Prototypes</em> which will be published as part of <a href="http://www.fdl-conference.org">FDL21</a>. The artifacts provided here use pre-compiled binaries and pre-generated stack usage databases. Based on these provided files, the stack size estimation and performance evaluation performed in the paper can be replicated. Various README.md files, which provide more information on individual artifacts, are also included.</p> <p>The software used in conjunction with these artifacts is also freely available on GitHub:</p> <ul> <li>https://github.com/agra-uni-bremen/fdl21-stackuse-vp</li> <li>https://github.com/agra-uni-bremen/stack-usage-db</li> </ul>
On the relation between technical debt indicators and quality criteria in Stack Overflow discussions
<p>This is the replication package for the paper "On the relation between technical debt indicators and quality criteria in Stack Overflow discussions". The paper was published on the Technical Research Track of the Bralizian Symposium on Software Engineering.This is the replication package for the paper "On the relation between technical debt indicators and quality criteria in Stack Overflow discussions". The paper was published on the Technical Research Track of the Bralizian Symposium on Software Engineering.</p>
Is reputation on Stack Overflow always a good indicator for users' expertise? No!
<p>Dataset for RQ1 and RQ2. </p>
Supplemental materials for studying GPU programming with Stack Overflow posts
<p>It includes the data used in our study of GPU programming and the complete results obtained from the study.</p>
GPT vs Stack Overflow: data collection (A2I2 T2 2023)
<p><strong>About</strong></p> <p>The dataset components produced by <a href="https://github.com/MHLoppy/A2I2-T2-2023">this repo</a>. Please see the documentation there for more information.</p> <p>Each CSV has been individually zipped so that you only have to download the specific file(s) that you want.</p> <p> </p> <p><strong>Overview of Files</strong></p> <p>From using the <a href="https://archive.org/details/stackexchange">Stack Exchange Data Dump</a> as the data source (these zip files have a <strong>DD_</strong> prefix):</p> <ul> <li>Raw dataset before processing: <strong>saved_dataset.csv (DD_saved_dataset.zip)</strong></li> <li>Completed tag count: <strong>tag_count.csv (DD_tag_count.zip)</strong></li> <li>Processed dataset with completed evaluations: <strong>dataset_results.csv (DD_dataset_results.zip)</strong></li> </ul> <p>From using Google BigQuery as the data source (these zip files have a <strong>BQ_</strong> prefix):</p> <ul> <li>Raw dataset before processing: <strong>saved_dataset.csv (BQ_saved_dataset.zip)</strong></li> <li>Completed tag count: <strong>tag_count.csv (BQ_tag_count.zip)</strong></li> <li><em>No large-scale evaluation was completed when using BigQuery as a data source.</em></li> </ul> <p>As noted in the linked repo, the use of Google BigQuery as a data source is not recommended for this work, but the working code and dataset have nonetheless been provided for completeness.</p> <p> </p> <p><strong>License</strong></p> <p>This dataset is licensed under the <a href="https://creativecommons.org/licenses/by-sa/4.0/">CC BY-SA 4.0 license</a>, the same license used by the Stack Exchange Data Dump.</p>
NLP2TestableCode Filtered Stack Overflow Dataset
<p>This dataset is for use in NLP2TestableCode, XML files containing Stack Overflow questions and answers tagged with Java. This dataset comes from a March 2019 upload of SO data from https://archive.org/details/stackexchange. Please extract and place in the data folder. The uncompressed dataset size is 6.84GB.</p>
Replication Package for the Paper: "How Do Users Revise Architectural Related Questions on Stack Overflow: An Empirical Study"
<p>This is the replication package for the paper: "How Do Users Revise Architectural Related Questions on Stack Overflow: An Empirical Study". In the following, we provide a brief description of the folders and files:</p> <p><strong>(1) raw data</strong></p> <p>The raw data folder contains the retrieved 36,417 posts and the SQL query used for retrieving ARPs from Stack Overflow through the query interface provided by Stack Exchange.</p> <p><strong>(2) filtered ARPs</strong></p> <p>The filtered ARPs folder contains 13,205 filtered candidates ARPs from the retrieved 36,417 posts and the results of data analysis for the first RQ (i.e., RQ1).</p> <p><strong>(3) randomly selected posts and labeling results</strong></p> <p>The randomly selected posts and labeling results folder contains 1,068 randomly selected posts and their labeling results (i.e., 21 ARPs, wherein 14.3%, 3 out of 21 ARPs, do not contain “architect*” terms and 85.7%, 18 out of 21 APRs, contain “architect*” terms).</p> <p><strong>(4) relevant ARPs for answering RQs</strong></p> <p>The relevant ARPs for answering RQs folder contains 4,114 ARPs with revision information for answering the last three RQs (i.e., RQ2, RQ3, and RQ4).</p> <p><strong>(5) interview responses</strong></p> <p>The interview responses folder contains 11 collected interview responses from software practitioners. These responses were gathered to evaluate the identified categories related to ARQ revisions.</p> <p><strong>(6) data extraction and analysis</strong></p> <p>The data extraction and analysis folder contains the MAXQDA file. Data Labeling & Encoding for RQs.mx20 is the results of data labeling and encoding for RQ2, RQ3, and RQ4, which were analyzed by the MAXQDA tool. This file can be opened by MAXQDA 2020 or higher versions, which are available at https://www.maxqda.com/ for download. You may also use the free 14 days trial version of MAXQDA 2020, which is available at https://www.maxqda.com/trial for download.</p>
Stack Overflow Duplicate Post Dataset
<p>As a part of the supplimentary material for the paper "Refining GPT-3 Embeddings with a Siamese Structure for Technical Post Duplicate Detection".</p> <p>dup_post_csv.tar.gz: The dataset contains all duplicate post pairs from Stack Overflow up to December 2022, with an 80%/20% split between the training and test sets.</p> <p>CQADupStack.tar: The CQADupStack Benchmark dataset. The package contains training and test sets from nine sub-domains. GPT-3 embeddings for all posts are appended.</p>
How solution snippets are presented in answers posted on Stack Overflow and how they could be potentially reused.
<p>Software engineering Q&A websites (e.g., Stack Overflow), harness the collective expertise of users to address technical queries. Over time, these platforms evolve into valuable repositories of software engineering knowledge. Such repositories serve as essential resources for developers looking for <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">solutions</abbr></abbr></abbr> to common programming <abbr title="Sentiment effect: -2">problems</abbr>. In Stack Overflow, developers may approach answering questions in various ways. <abbr title="Sentiment effect: 2">Gaining</abbr> insight into how developers formulate their answers on Stack Overflow can enhance knowledge sharing and streamline the process of finding <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">solutions</abbr></abbr></abbr>. Furthermore, such insights could also inform improvements in Generative Artificial Intelligence (GenAI) tools to <abbr title="Sentiment effect: 2">better</abbr> align generated source code for comprehension and understandability, as AI-generated answers are known to include irrelevant information and hallucinations. In this study, we seek to deepen the understanding of how <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">solutions</abbr></abbr></abbr> are presented on Stack Overflow. We conducted an empirical study that investigates programming questions that are answered with a <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">Solution</abbr></abbr></abbr></abbr></abbr> Snippet to understand how a <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">Solution</abbr></abbr></abbr></abbr></abbr> Snippet is presented, and the ways how it should be adapted when it is reused. Our study resulted in two categorizations: 1) eight categories of how <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">Solution</abbr></abbr></abbr></abbr></abbr> Snippets are presented on Stack Overflow answers and 2) five categories of how <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">Solution</abbr></abbr></abbr></abbr></abbr> Snippets could be adapted for reuse. Then, we analyzed these categorizations and discussed the implications. We anticipate that Stack Overflow will remain a valuable resource for the foreseeable future, and the insights revealed in our paper lay the groundwork for <abbr title="Sentiment effect: 2">improving</abbr> program comprehension of <abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1"><abbr title="Sentiment effect: 1">Solution</abbr></abbr></abbr></abbr></abbr> Snippets on Stack Overflow and GenAI tools.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.