The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Chinese Patents from 1985 to 2025
(ChinesePatentsFrom1985to2025)
Dataset Summary
Chinese Patents from 1985 to 2025 (ChinesePatentsFrom1985to2025) is a large-scale structured dataset constructed from publicly available Chinese patent documents published between 1985 and 2025.2.
The dataset is intended to support academic research and educational use in areas including patent analysis, innovation studies, natural language processing, and knowledge-based engineering design. It emphasizes preserving original technical descriptions while enabling structured analysis for downstream research tasks.
Supported Tasks and Applications
This dataset can be used for a wide range of research purposes, including but not limited to:
Patent text classification
Functional and effect-oriented text analysis
Patent-based knowledge graph construction
Innovation pattern mining
Scientific and technical information extraction
Design-by-analogy and creativity support research
Chinese NLP tasks (e.g., segmentation, NER, relation extraction)
No official leaderboard is associated with this dataset.
Languages
Chinese (Simplified)
Dataset Structure
Each entry in the dataset corresponds to a patent record derived from official public disclosures. Depending on the release version, records may include:
Patent title
Abstract
Application year
Note: Full patent claims are not guaranteed to be included unless explicitly stated for a given release.
Data Sources
The dataset is constructed from publicly available patent documents released by:
China National Intellectual Property Administration (CNIPA)
All original patent texts remain the intellectual property of their respective rights holders. This dataset represents a secondary organization and structuring of publicly disclosed information.
Data Collection and Processing
Patent documents were collected from official public disclosure channels.
Text data were cleaned and normalized to remove formatting noise.
Metadata fields were standardized to ensure temporal and semantic consistency.
No private, confidential, or non-public information was included.
Intended Uses Primary Intended Uses
Academic research
Teaching and coursework
Methodological validation
Non-commercial experimentation
Benchmarking NLP and knowledge representation methods
Out-of-Scope Uses
Direct commercial exploitation
Deployment in commercial AI products or services
Large-scale proprietary model training without explicit permission
License
This dataset is released under the OpenRAIL License.
While the underlying patent documents are publicly available, the curation, structuring, and organization of this dataset constitute a separate scholarly contribution.
Users are required to:
Provide proper attribution when using the dataset
Respect restrictions on commercial usage
Ensure compliance with applicable laws and intellectual property regulations
Citation
If you use this dataset in academic work, please cite it as follows:
@dataset{liu_chinesepatents_1985_2025, title = {Chinese Patents from 1985 to 2025}, author = {Liu, Hongwei}, year = {2025}, version = {v2025.02}, publisher = {Hugging Face}, url = {https://huggingface.co/datasets/LiuHongwei1992/ChinesePatentsFrom1985to2025} }
Ethical Considerations
The dataset contains only publicly disclosed patent information.
No sensitive personal data beyond legally published patent content is intentionally included.
Users are responsible for ensuring ethical and lawful use of the dataset.
Limitations
Patent texts may contain domain-specific terminology requiring expert interpretation.
Automatic processing may introduce minor normalization or segmentation errors.
This dataset does not replace official patent databases for legal or administrative purposes.
Versioning and Maintenance
The dataset follows a versioned release policy to ensure reproducibility and long-term usability.
Example version format:
v2025.02 v2025.02.1 v2025.03.0
Contact
For questions, feedback, or academic collaboration, please contact:
Hongwei Liu (scu_liuhongwei@163.com;liuhongwei@suit-sz.edu.cn) Dataset Maintainer
- Downloads last month
- 68