Loading...
Loading...
Social network graphs, knowledge graphs, citation networks, molecular graphs for GNN, web link graphs
428 datasets
A dataset of graphs likely enriched with features for training and evaluating Temporal Graph Neural Networks (TGNNs). The dataset is hosted on Kaggle, but its specific size, origin, and creation date are not provided in the available metadata. The title suggests the graphs are processed or augmented to support advanced graph learning tasks.
An identity graph dataset published on Kaggle by Hamed Behrouzi. The dataset likely contains nodes and edges representing entities and their relationships, potentially for linking user identities across platforms. Specific details on size, time range, and collection method are not provided in the metadata.
CAGNN Sum Checkpoints Wajahatb v1 is a dataset of pre-trained model checkpoints published on Kaggle. The title suggests it relates to a Graph Neural Network (GNN) architecture named CAGNN. The dataset likely contains saved model states for tasks involving graph-structured data.
CAGNN Sum v6 Checkpoints are pre-trained model files hosted on Kaggle. The title suggests the data relates to a Graph Neural Network architecture, likely for tasks involving graph-structured data. Metadata is minimal; actual content and performance require verification after download.
A graph dataset representing connections between artist pages on the Facebook social media platform. The dataset's exact size, creator, and creation date are not specified in the provided metadata. It was sourced from Kaggle, a platform for sharing datasets.
CAGNN Sum Checkpoints Wajahatb is a collection of pre-trained model checkpoints for a Graph Neural Network (GNN) architecture, likely the Context-Aware Graph Neural Network (CAGNN). The dataset contains saved model states from the Kaggle user 'Wajahatb', intended for tasks involving a 'sum' operation on graph-structured data. Specific details on the number of checkpoints, training data, and creation date are unavailable.
Harvard Dataverse hosts a dataset by YUHAO WANG analyzing spatial-institutional misfit within China's Marine Protected Area network. The dataset supports research in Social Sciences and Earth and Environmental Sciences, though specific row and column counts are not provided. It was last updated in January 2026.
19,428 nodes representing Marvel characters and 12,942 comic book issues connected by 96,662 edges. The data includes a bipartite graph of hero-comic appearances and a projected social network of hero-to-hero interactions.
150,101 English lexemes (words or multi-word expressions) are included in this synthetic encyclopedic dictionary and semantic knowledge graph. It integrates lexicographic definitions, encyclopedic context, etymological histories, and semantic relationships. The dataset was created by author mjbommar and was last updated on 2025-11-23.
536,829 sense definitions across 150,101 English lexemes are provided in this synthetic encyclopedic dictionary and semantic knowledge graph. The dataset integrates lexicographic definitions, encyclopedic context, etymological histories, and semantic relationships in a unified resource. It was authored by mjbommar and last updated on 2025-11-23.
199,998 Indonesian legal regulations are available in a SQLite database version of a knowledge graph, optimized for querying. The database includes 199,998 embeddings and 199,998 TF-IDF vectors, totaling 2273.46 MB. It was created by Azzindani and last updated on November 3, 2025.
The 3Gpp Knowledgegraph dataset provides a Knowledge Graph representation of 3GPP Release 13 documentation, compiled by Khalifa University. It contains 12,800 nodes and 9,970 edges extracted from the official 3GPP site. The dataset was uploaded to Hugging Face by otellm on November 24, 2025.
ConceptNet 5.7 is a common-sense knowledge graph containing approximately 8 million entities and RDF triples of everyday facts and relationships. It is converted to HuggingFace dataset format for machine learning use, with an extracted size of 1.2 GB.
Social Network First? is a research project by the Empowerment & Professionalisation research group of Inholland University of Applied Sciences, the research bureau of HVO-Querido, and the Urban Social Work research group of the Amsterdam University of Applied Sciences. The project investigated how social service organizations in Amsterdam and Haarlem work to support the social networks of people experiencing homelessness over a two-year period. The dataset was authored by B. van der Ent and is archived by DANS Data Station Social Sciences and Humanities.
A benchmark for evaluating the faithfulness of large language models, built on a financial knowledge graph. It contains over 480,000 triplet evaluations sourced from 101 S&P 100 companies, including judge reasoning and full document context. The dataset was curated by Domyn and was last updated on October 6, 2025.
S&P 500 companies' 10-K filings from 2014 to 2024 provide the source for this financial knowledge graph. Domyn curated FinReflectKG, which contains 17.51 million normalized triplets with textual context. The dataset was last updated on Hugging Face in October 2025.
Graph datasets derived from Finite Element Method (FEM) simulations performed in FEBio for two benchmark problems. The data is intended for training and testing the performance of graph neural network architectures. Author Vijay Dubey contributed this dataset to the Texas Data Repository, with a last update recorded on October 15,我们发现了一个错误。
3D finite element meshes and stress distributions generated for stochastic point elastic loading conditions by cmudrc in 2025. This mechanical engineering dataset supports the training of graph neural network surrogate models to accelerate stochastic finite element method (SFEM) simulations.
T-GRAB is a synthetic reasoning benchmark designed for learning on temporal graphs. It was created by Gilestel and the dataset entry was last updated on August 19, 2025. The benchmark is associated with a research paper accessible via the provided URL.
Constructed from Wikipedia and Wikidata, this knowledge graph contains approximately 4.6 million entities and 21 million triples. The dataset was created by Alphonse7 and was last updated on June 22, 2025.