Update README.md

README.md

@@ -1,3 +1,51 @@
----
-license: cc-by-sa-4.0
----
+---
+license: cc-by-sa-4.0
+---
+
+This dataset accompanies our ACL 2025 paper: [Internal and External Impacts of Natural Language Processing Papers](https://arxiv.org/abs/2505.16061).
+
+We present a comprehensive dataset for analyzing both the internal (academic) and external (public) impacts of NLP papers published in top-tier conferences — ACL, EMNLP, and NAACL — between 1979 and 2024. The dataset supports a wide range of scientometric studies, including topic-level impact evaluation across patents, media, policy, and code repositories.
+
+## Data Sources
+Our dataset integrates signals from several open and restricted resources:
+
+- [ACL Anthology](https://github.com/acl-org/acl-anthology): NLP papers
+
+- [OpenAlex](https://docs.openalex.org/download-all-data/download-to-your-machine): Citation counts
+
+- [Reliance on Science](https://zenodo.org/records/11461587): Patent-to-paper links.
+
+- [Papers with Code](https://production-media.paperswithcode.com/about/links-between-papers-and-code.json.gz) + GitHub API: Linking NLP papers to GitHub repositories, including stars and forks.
+
+⚠️ **Not publicly included**:
+
+Altmetric (media-to-paper links) and Overton (policy-document-to-paper links) are used in our analysis but are not released here due to data access restrictions. Approval from Altmetric and Overton needed to access these signals. Please refer to our paper for more details.
+
+## Dataset Format
+Each record corresponds to one NLP paper and includes the following fields:
+
+| Field Name         | Description                                                                 |
+|--------------------|-----------------------------------------------------------------------------|
+| `title`            | Title of the paper                                                          |
+| `abstract`         | Abstract of the paper                                                       |
+| `doi`              | Digital Object Identifier(s)                                                   |
+| `venue`            | Conference venue (ACL, EMNLP, or NAACL)                                        |
+| `year`             | Year of publication                                                         |
+| `oaid`             | OpenAlex ID(s)                                                                 |
+| `label`            | Assigned research topic (based on ACL 2025 Call for Papers; predicted via GPT-4o) |
+| `paper_citation`   | Citation count (OpenAlex)                                                |
+| `patent_citation`  | Number of times cited in USPTO patents (Reliance on Science)               |
+| `github_repo`      | List of associated GitHub repository URLs (Papers with Code)            |
+| `github_star`      | Total number of GitHub stars for associated repositories                    |
+| `github_fork`      | Total number of GitHub forks for associated repositories                    |
+
+## Citation
+If you find this dataset useful, please cite the following paper:
+```
+@article{zhang2025internal,
+  title={Internal and External Impacts of Natural Language Processing Papers},
+  author={Zhang, Yu},
+  journal={arXiv preprint arXiv:2505.16061},
+  year={2025}
+}
+```