Hugging Face's logo

Datasets:
UEXdo
/
OpenDataHub

Tasks:
Text Classification
Tabular Classification
Size:
1K<n<10K
Tags:
energy
urban-energy
carbon-neutrality
climate-change
sustainability
open-data
License:
Dataset card Files
xet
 Community
OpenDataHub / README.md
DengVvi's picture
DengVvi
Update README.md
8ccae1a verified
preview code
|
raw
history blame contribute delete
7.92 kB
metadata 
dataset_info:
  features:
    - name: title
      dtype: string
    - name: doi
      dtype: string
    - name: url
      dtype: string
    - name: journal
      dtype: string
    - name: year
      dtype: int32
    - name: authors
      dtype: string
    - name: abstract
      dtype: string
    - name: direction
      dtype: string
    - name: subcategory
      dtype: string
  splits:
    - name: train
      num_bytes: 1891581
      num_examples: 2056
  download_size: 500000
  dataset_size: 1891581
configs:
  - config_name: default
    data_files:
      - split: train
        path: datasets_classified.json
license: cc-by-4.0
task_categories:
  - text-classification
  - tabular-classification
tags:
  - energy
  - urban-energy
  - carbon-neutrality
  - climate-change
  - sustainability
  - open-data
  - nature
  - science
  - cell
pretty_name: UEX Open Datasets Database
size_categories:
  - 1K<n<10K

UEX Open Datahub

Dataset Description

Datahub Summary

The UEX Open Datahub is a curated collection of metadata for open datasets published in top-tier academic journals (Nature, Science, and Cell series) from 2020 to the present. It is maintained by the International Centre of Urban Energy Nexus (UEX) at The Hong Kong Polytechnic University (PolyU).

The database is specifically designed to support research in urban energy and environmental challenges, accelerating the transition to urban carbon neutrality. It categorizes over 2,000 dataset records into four core research directions aligned with the UEX Centre's laboratories:

  1. 🌱 CleanTech: Novel Low/Zero Carbon Technologies
  2. ⚡ FLEXERGY: Demand Response & New Mobilities & Urban Planning
  3. 🤖 DigiEnergy: AI & Data Science for Urban Energy Systems
  4. 💹 EnergiTrade: Carbon Trading & New Business Models

Supported Tasks and Leaderboards

This dataset is primarily intended for:

  • Literature Review & Meta-Analysis: Discovering high-quality open datasets in specific urban energy domains.
  • Data Discovery: Finding raw data for training machine learning models in energy forecasting, climate modeling, and materials science.
  • Trend Analysis: Analyzing research trends in carbon neutrality and urban energy systems across top-tier journals.

Languages

The metadata (titles, abstracts, authors) is in English (en).

Dataset Structure

Data Instances

A typical instance in the dataset represents a single academic paper that published an open dataset:

{
  "title": "UrbanEV: An Open Benchmark Dataset for Urban Electric Vehicle Charging Demand Prediction",
  "doi": "10.1038/s41597-025-04874-4",
  "url": "https://doi.org/10.1038/s41597-025-04874-4",
  "journal": "Scientific Data",
  "year": 2025,
  "authors": "Li, H.; Qu, H.; Tan, X.;You, L.; Zhu, R.; Fan, W.",
  "abstract": "The recent surge in electric vehicles (EVs), driven by a collective push to enhance global environmental sustainability...",
  "direction": "FLEXERGY",
  "subcategory": "Electric Vehicles & Mobility"
}

Data Fields

  • title (string): The title of the academic paper.
  • doi (string): The Digital Object Identifier of the paper.
  • url (string): The URL to access the paper (usually https://doi.org/{doi}).
  • journal (string): The name of the journal where the paper was published.
  • year (int32): The publication year (2020 or later).
  • authors (string): A semicolon-separated list of authors.
  • abstract (string): The abstract of the paper.
  • direction (string): The primary UEX research direction (e.g., CleanTech, FLEXERGY).
  • subcategory (string): The specific sub-domain within the research direction.

Data Splits

The dataset contains a single split: train.

Dataset Creation

Curation Rationale

The transition to carbon-neutral cities requires vast amounts of high-quality data. While top-tier journals increasingly mandate open data, these datasets are scattered across various repositories and publications. This database aggregates and categorizes these datasets specifically for urban energy researchers, saving hundreds of hours of literature search.

Source Data

Initial Data Collection and Normalization

Data was collected programmatically using the CrossRef API. The collection process targeted specific ISSNs of 13 journals across three major publishers:

  • Nature Portfolio: Scientific Data, Nature Energy, Nature Sustainability, Nature Communications, Nature Climate Change, npj Clean Energy, Nature Cities, Scientific Reports
  • Cell Press: Joule, One Earth, Cell Reports Sustainability, iScience
  • AAAS: Science Advances

The search was restricted to publications from 2020 to the present that contain keywords related to energy, carbon, climate, and urban systems.

Classification Methodology

The collected records were filtered for relevance and classified using a multi-level, rule-based keyword matching algorithm. The classification maps each paper to one of the four UEX research directions and a specific subcategory based on the presence of domain-specific terminology in the title and abstract.

Annotations

The dataset does not contain manual annotations. The direction and subcategory labels are automatically generated based on predefined keyword rules aligned with the UEX Centre's research taxonomy.

Personal and Sensitive Information

The dataset contains only publicly available bibliographic metadata. It does not contain personal or sensitive information.

Considerations for Using the Data

Social Impact of Dataset

By centralizing access to high-quality datasets on renewable energy, carbon trading, and urban sustainability, this database aims to accelerate research that mitigates climate change and promotes sustainable urban development.

Discussion of Biases

  • Source Bias: The dataset only includes papers from selected high-impact journals (Nature, Science, Cell series). It does not cover datasets published in specialized domain journals (e.g., IEEE, Elsevier energy journals) or standalone data repositories without an accompanying paper.
  • Language Bias: The search and classification rely entirely on English keywords, excluding datasets published in other languages.
  • Classification Limitations: The automated keyword-based classification may occasionally misclassify papers, especially those spanning multiple interdisciplinary domains.

Additional Information

Dataset Curators

This dataset was curated by the International Centre of Urban Energy Nexus (UEX) at The Hong Kong Polytechnic University.

Licensing Information

The metadata in this repository is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Note: The actual datasets linked within this database are subject to their own respective licenses as defined by the original authors and publishers.

Citation Information

If you use this database in your research, please cite it as follows:

@misc{uex_open_datasets,
  title        = {{UEX Open Datasets Database: Curated Open Datasets for Urban Energy Research}},
  author       = {{International Centre of Urban Energy Nexus (UEX), PolyU}},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/datasets/UEXdo/Public_Datasets}},
  note         = {The Hong Kong Polytechnic University}
}

Contributions

Thanks to the UEX Research Centre team for defining the research taxonomy and supporting the creation of this open-source initiative.