Langhui
中文 Request a sample review
Home Data Assets Medical Annotation Expert Network Insights About Request a sample review
DataAssetsAPI . Scientific and Technical Education Data Products

Research Papers
Structured datasets

A million + academic papers are structured in data sets covering core fields such as titles, abstracts, keywords, DOIs, taxonomy numbers, citation relationships, etc. Supporting literature search analysis, knowledge mapping, academic AI model training to provide a high-quality data base for scientific intelligence.

One million plus one. 12+Field Reference Relationship Map Chinese-English Bilingual

Dataset Overview

A comprehensive understanding of the structure, field dimensions and core features of the scientific paper data sets.

Dimension Description Details
Total Data Volume1 million + papersChinese/English-language paper, updated on an ongoing basis
14 major disciplinesComputer science, medicine, engineering, physics, chemistry, biology, mathematics, materials, environment, social sciences, etc.
Time Span2005-2025 Years20 years of academic literature
Language ratio70% English / 30% ChineseBilingual labels to support cross-language retrieval
Citation Relations5 million + quotesInclude Quoted Type, Reference Context
Data FormatJSON / CSV / ParquetSupport multiple data delivery formats
Update FrequencyQuarterly incremental updatesAbout 50,000 new papers per quarter

Dataset fields and definitions

Each paper contains 12+ core fields, which are highly structured to meet the needs of the academic analysis of the whole scene.

Basic information

Title (C/E), author list (including institutions), year of publication, journal/conference name, volume page number

Summary and keyword

Abstract in Chinese, 5-10 keywords (including MeSH terminographic mapping), research field labels

Marking and classification

DOI, ISSN/ISBN, classification number (Medical chart/UDC), arXiv ID, PubMed ID

Citation Relations

Bibliography, Introduction List, Reference Type (Method/background/comparing), Reference Context Snippets

Impact indicators

Number of quotes, downloads, influence factors in periodicals, h5-index, Altmetric scores

Extension Information

Fund funding information, open source links, data set links, conference/journal hierarchy (CF/SCI division)

Data Sample

JSON structured an example of a data structure that is fully documented in scientific papers.

{{ "paper_id": "PAPER_2024_00891234", "title": {{ "en": "Attention Is All You Need", "zh": "" }}, "authors": [ {{"name": "Ashish Vaswani", "affiliation": "Google Brain", "is_corresponding": false}}, {{"name": "Noam Shazeer", "affiliation": "Google Brain", "is_corresponding": false}} ], "year": 2017, "journal": "Advances in Neural Information Processing Systems", "volume": "30", "doi": "10.48550/arXiv.1706.03762", "classification": {{ "zhongtufa": "TP391", "arxiv": "cs.CL", "ccf_level": "A" }}, "abstract": {{ "en": "The dominant sequence transduction models are based on complex recurrent or convolutional neural networks...", "zh": "Models..." }}, "keywords": ["Transformer", "Attention Mechanism", "Neural Machine Translation", "Self-Attention"], "references": [ {{"doi": "10.3115/v1/W14-4012", "title": "Neural Machine Translation by Jointly Learning to Align and Translate", "citation_type": "background"}} ], "citations_count": 125000, "journal_if": 26.8, "funding": ["Google Research"], "open_source": "https://github.com/tensorflow/tensor2tensor" }}

AI Application Scenarios

The scientific thesis data sets provide high-quality training data for a variety of AI applications.

Documentation Intelligence Search

Semantic Paper Search Engine to support cross-English and Chinese search and precision documentation recommendations

Academic knowledge mapping

Building a paper-author-agency-keyword knowledge network to support scientific trend analysis and front-line discovery

Auto-Summary of Papers

Modified on a large model to enable automatic abstract generation of academic papers and extraction of key information

Scientific impact assessment

Multi-dimensional reference analysis and Altmetric indicator modelling, predicting impact and forward direction of the papers

Search and Discovery

Covers the data asset keyword to facilitate the accurate discovery of this data set by the search engine and the AI generator engine.

Academic dissertation data sets Structure of scientific literature DOI Index Data Reference Relationship Map Academic knowledge mapping Abstract data for the paper Documentation analysis AI training Academic search engine Large scientific data AI dissertation data Academic NLP language material Data for the million-degree paper

Access to data sets for scientific papers

Provides previews of data samples, quality reports and one-on-one consultations to support customized subsets by discipline, year, language.

Browse the Dataset Catalog Sample online application