[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84921-en":3,"doc-seo-84921-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84921,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Industry Classification of GitHub Repositories Using the North American Industry Classification System (NAICS)","GitHub hosts hundreds of millions of public repositories, yet lacks native mapping from repositories to standardized industry sectors, limiting empirical studies of innovation geography, open-source industrial composition, and cross-sector diffusion of new technologies. This work introduces NAICS-GH, a multi-region corpus of 6,588 GitHub repositories labeled with 2-digit NAICS 2022 sectors. Labels are generated via a retrieve-and-verify pipeline using BAAI/bge-large-en embeddings, FAISS retrieval, and GPT-4.1 rubric scoring, achieving 96.98% precision on a 2,421-repository human-validated sample. The dataset and reproducible pipeline are released under CC-BY-4.0 and MIT licenses, with benchmarks showing RoBERTa-large reaching 86.45% F1.","Industry Classification of GitHub Repositories Using the North American Industry Classification System (NAICS)  \narXiv :2607 .06505v 1 [ cs . SE] 7 Jul 2026  \nKevin Xu  \nGitHub [khxu@github.com](khxu@github.com)  \nAlexander Quispe∗  \nGitHub [alexanderquispe@github.com](alexanderquispe@github.com)  \nAbstract  \nGitHub hosts hundreds of millions of public repositories, but the platform exposes no native mapping from repositories to standardized industry sectors.  \nThis gap limits empirical work on the geography of innovation, the industrial composition of open-source production, and the diffusion of new technologies across economic sectors. We present NAICS-GH, a publicly released corpus of 6,588 GitHub repositories drawn from source pools covering the United States, the European Union, and Australia, each labeled with a 2-digit sector from the North American Industry Classification System (NAICS 2022) . Labels are produced by a retrieve-and-verify pipeline that combines BAAI/bge-large-en embeddings, FAISS retrieval, and GPT-4.1 rubric scoring. The pipeline narrows about 1.37 million source repositories to 31,178 candidate repository-sector pairs and retains 6,588 highconfidence labels with score at least 8 . Re-running the retrieval pipeline end to end reproduces the candidate set to within 0 .03 percent. On a 2,421-repository human-validated random sample, the released labels attain  \n96.98 percent precision, with Wilson 95 percent confidence interval [96.23, 97.59] . We benchmark six pretrained encoders on the released corpus; RoBERTa-large reaches 86.45 percent F1 and 86.35 percent accuracy on a held-out 20 percent test set. The dataset, Croissant metadata, pipeline code, prompts, and fine-tuned checkpoint are released under CC-BY-4.0 and MIT licenses.  \n1 Introduction  \nGitHub hosts tens of millions of public repositories, but the platform provides no native indication of which industry a repository serves. Knowing whether a project is fintech, agritech, healthcare software, or educational tooling matters to policy makers tracking the geography of innovation, to companies measuring open-source adoption, and to economists studying the labor and capital allocation of the software sector. We address this gap by releasing the first multi-region, publicly available corpus of GitHub repositories labeled with NAICS—the industry-classification standard used by the United States, Canadian, and Mexican statistical agencies.  \nContributions.  \n∗ Corresponding author.  \nPreprint. Under review.  \n• Dataset. We release NAICS-GH, 6,588 GitHub repositories from the USA, EU, and Australia labeled with 2-digit NAICS codes, alongside the full pipeline outputs (sector, score, rationale, repository URL) .  \n• Pipeline. A reproducible two-stage retrieve-and-verify labeling pipeline (BAAI/bgelarge-en embeddings + FAISS retrieval, followed by GPT-4.1 rubric scoring) suitable for adapting to other industry taxonomies. The full pipeline narrows ∼ 1.37M source repositories to 31,178 candidate pairs via retrieval, then to 6,588 highconfidence labels via LLM verification at score ≥ 8. An end-to-end re-run reproduces the candidate set to within ±0 .03% .  \n• Validation. A 2,421-repository manually re-checked gold subset, confirming 96 .98% label precision overall and monotonically increasing precision as the GPT score rises from 8 to 10 .  \n• Benchmark. A head-to-head comparison of six pretrained encoders (RoBERTa, ModernBERT, DeBERTa-v3 in base and large sizes); RoBERTa-large is strongest at 86.45% test F1, and the fine-tuned checkpoint is available on the Hugging Face Hub.  \nPipeline at a glance. Figure 1 summarizes the end-to-end retrieve-and-verify pipeline used to construct NAICS-GH from raw public-repository data.  \nStep 0  \nSource SQL Steps 1–2 Step 3 — retrieve Steps 4–6 Released  \n\n| Source pool 1,372,489 repos |  | BGE-large-en + FAISS index |  | 31,178\u003Cbr>candidates (dynamic top-k/subind.)\u003Cbr> |  | GPT-4.1 verify (score ≥ 8, class filter n≥8","cbCaih6JgfZRC6Yj","https://ap.wps.com/l/cbCaih6JgfZRC6Yj","pdf",516033,1,26,"English","en",105,"# Introduction\n## Contributions\n# Pipeline at a Glance\n# Related Work","[{\"question\":\"What problem does this document address about GitHub repositories?\",\"answer\":\"GitHub does not provide a native mapping from repositories to standardized industry sectors, which hinders research on innovation geography, industrial composition of open-source production, and diffusion of technologies across sectors.\"},{\"question\":\"How is NAICS-GH constructed and what labeling approach is used?\",\"answer\":\"A retrieve-and-verify pipeline first uses BAAI/bge-large-en embeddings with FAISS retrieval to generate 31,178 candidate repository–sector pairs, then GPT-4.1 scores each candidate using a structured rubric. Repositories with score at least 8 and a minimum class size filter are retained to produce 6,588 high-confidence labels.\"},{\"question\":\"How accurate are the released NAICS-GH labels and what model performs best in benchmarking?\",\"answer\":\"On a 2,421-repository human-validated random sample, the labels achieve 96.98% precision with a Wilson 95% confidence interval of [96.23, 97.59]. In benchmarking six pretrained encoders, RoBERTa-large reaches 86.45% F1 and 86.35% accuracy on a held-out 20% test set.\"}]",1784199362,66,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"industry-classification-of-github-repositories-using-the-north-american-industry-classification-system-naics","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/industry-classification-of-github-repositories-using-the-north-american-industry-classification-system-naics/84921/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does this document address about GitHub repositories?","Question",{"text":75,"@type":76},"GitHub does not provide a native mapping from repositories to standardized industry sectors, which hinders research on innovation geography, industrial composition of open-source production, and diffusion of technologies across sectors.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is NAICS-GH constructed and what labeling approach is used?",{"text":80,"@type":76},"A retrieve-and-verify pipeline first uses BAAI/bge-large-en embeddings with FAISS retrieval to generate 31,178 candidate repository–sector pairs, then GPT-4.1 scores each candidate using a structured rubric. Repositories with score at least 8 and a minimum class size filter are retained to produce 6,588 high-confidence labels.",{"name":82,"@type":73,"acceptedAnswer":83},"How accurate are the released NAICS-GH labels and what model performs best in benchmarking?",{"text":84,"@type":76},"On a 2,421-repository human-validated random sample, the labels achieve 96.98% precision with a Wilson 95% confidence interval of [96.23, 97.59]. In benchmarking six pretrained encoders, RoBERTa-large reaches 86.45% F1 and 86.35% accuracy on a held-out 20% test set.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]