Knowledge Tree
Wikidata
Facts (40)
Sources
Construction of Knowledge Graphs: State and Challenges - arXiv arxiv.org 22 facts
referencePintscher authored 'From freebase to wikidata: The great migration', which was published in the Proceedings of the 25th international conference on world wide web in 2016, pages 1419–1428.
referenceYAGO 4 constructs its knowledge graph by collecting data from Wikidata and forcing it into a taxonomy based on schema.org and Bioschemas, rather than using Wikipedia, WordNet, and GeoNames data.
procedureExternal knowledge bases can be queried to enhance knowledge graph data by using extracted global persistent identifiers (PID) such as ISBN numbers, DOIs, or ORCIDs to request information from Wikidata.
measurementThe Artificial Intelligence Knowledge Graph (AI-KG) contains over 820,000 entities derived from over 333,000 research publications, integrating data from the Microsoft Academic Graph, the Computer Science Ontology, and Wikidata.
claimFreebase, Wikidata, and the Open Research Knowledge Graph (ORKG) are manually curated knowledge graphs based on crowdsourcing.
procedureThe dstlr framework enriches its knowledge graph by manually defining mappings between Stanford CoreNLP relations and Wikidata properties to extract facts from external knowledge graphs.
measurementWikidata is the largest open-source knowledge graph, containing approximately 100 million entities with 300,000 entity types and 14 billion relations with 300,000 relation types.
referenceFrey et al. presented 'DBpedia FlexiFusion the best of Wikipedia> Wikidata> your data' at the International Semantic Web Conference in 2019.
referenceThe dstlr tool extracts mentions and relations from text, links them to Wikidata, and populates the resulting knowledge graph with additional facts from Wikidata.
claimFreebase was one of the first collaboratively built and versioned knowledge graphs and became a popular source for building other knowledge graphs, such as Wikidata, after its shutdown in 2016.
measurementThe Wikidata knowledge graph, established in 2012, contains 100 million entities and 14 billion facts, utilizing Custom and RDF formats.
procedureThe YAGO 4 taxonomy enrichment process involves mapping Wikidata to 235 schema.org classes and 116 relations, filtering low-coverage entities, and accepting entities transitively connected to initial classes via sub-class relations to create a final taxonomy of 10k classes.
claimThe dstlr tool links entities to Wikidata and fetches stored properties from this external source for knowledge completion.
claimWikidata facilitates semi-automatic curation involving both bots and human curators.
referenceA. Piscopo, L.-A. Kaffee, C. Phethean, and E. Simperl authored 'Provenance information in a collaborative knowledge graph: an evaluation of Wikidata external references', published in the International semantic web conference proceedings by Springer in 2017, pages 542–558.
claimYAGO maintains fact provenance by utilizing Wikidata’s annotations for validity time and external references.
procedureThe dstlr framework uses Stanford CoreNLP to perform named-entity extraction, relation extraction, and linking of extracted mentions to Wikidata.
claimThe Wikimedia Foundation releases full data dump snapshots of Wikidata twice a month.
claimYAGO, DBpedia, NELL, and Wikidata are examples of open knowledge graphs.
measurementDBpedia performs monthly batch extractions consuming data from up to 140 wikis, including various language-specific Wikipedia versions, Wikidata, and Wikimedia Commons, with each release undergoing completeness and quality validation.
procedureThe WorldKG ontology construction process involves fetching geographic class information tags from OpenStreetMap, using key-value pairs to infer a class hierarchy, and aligning these classes with Wikidata and DBpedia using an unsupervised machine learning approach followed by manual verification.
claimWikidata supports entity annotation with key-value pairs, including validity time, provenance, and references.
Large Language Models Meet Knowledge Graphs for Question ... arxiv.org Sep 22, 2025 8 facts
referenceThe EFSUM method, proposed by Ko et al. in 2024, performs KG Fact Summarization and uses KG Helpfulness and Faithfulness Filters with GPT-3.5-Turbo, Flan-T5-XL, and Llama-2-7B-Chat models and dataset-inherent knowledge graphs (Freebase, Wikidata) for KGQA and Multi-hop QA, evaluated using Accuracy (Acc) on the WQSP and Mintaka datasets.
referenceThe KnowLA method, proposed by Luo et al. in 2024, utilizes knowledgeable adaptation with Llama2-7B and Alpaca2 language models, incorporating WordNet, ConceptNet, and Wikidata knowledge graphs to perform MCQA, CBQA, and TruthfulQA tasks on the CSQA, SIQA, BBH, WQSP, and TriviaQA datasets, evaluated using Acc, CE Score, BLEU, and ROUGE metrics.
referenceLongRAG, as described by Zhao et al. (2024a), utilizes domain-specific fine-tuning for RAG and CoT-guided filtering with models including ChatGLM3-6B, Qwen1.5-7B, Vicuna-v1.5-7B, Llama-3-8B, GPT-3.5-Turbo, and GLM-4, applied to Wikidata for KBQA and Multi-hop QA tasks.
referenceThe LPKG method, proposed by Wang et al. in 2024, involves Planning LLM Tuning, Inference, and Execution using GPT-3.5-Turbo, CodeQwen1.5-7B-Chat, and Llama-3-8B-Instruct models with dataset-inherent knowledge graphs (Wikidata) and Wikidata15K for KGQA and Multi-hop QA, evaluated using EM, P, and R metrics on the HotpotQA, 2WikiMQA, Bamboogle, MuSiQue, and CLQA-Wiki datasets.
referenceThe InteractiveKBQA method, proposed by Xiong et al. in 2024, uses Multi-turn Interaction for Observation and Thinking with GPT-4-Turbo, Mistral-7B, and Llama-2-13B models and Freebase, Wikidata, and Movie KG knowledge graphs for KBQA and domain-specific QA, evaluated using F1, Hits@1, EM, and Acc metrics on the WQSP, CWQ, KQA Pro, and MetaQA datasets.
referencePATQA (Meem et al., 2024) is a temporal question-answering dataset that provides Wikidata questions for present-anchored temporal question answering.
claimACT-Selection (Salnikov et al., 2023) filters and re-ranks answer candidates based on their types extracted from Wikidata.
referenceThe ACT-Selection method, proposed by Salnikov et al. in 2023, utilizes ACT-based Answer Selection and Ranking with the T5-Large-SSM language model and Wikidata knowledge graph to perform KGQA, CBQA, and Multi-lingual KGQA tasks, evaluated using the Hit@1 metric on the SQ, RuBQ, and Mintaka datasets.
EdinburghNLP/awesome-hallucination-detection - GitHub github.com 3 facts
referenceThe MultiHal benchmark is a factual language modeling benchmark that extends previous benchmarks such as Shroom2024, HaluEval, HaluBench, TruthfulQA, Felm, Defan, and SimpleQA by mining relevant knowledge graph paths from Wikidata.
referenceThe research paper modeling factual queries as constraint-satisfaction problems utilizes AUROC and risk-coverage curve operating points as evaluation metrics, and uses CounterFact and factual queries generated from Wikidata as datasets.
measurementEvaluation metrics for list-based questions on Wikidata and Wiki-Category List include test precision and the average number of positive and negative hallucination entities; MultiSpanQA uses F1, Precision, and Recall; and longform generation of biographies uses FactScore.
Combining Knowledge Graphs and Large Language Models - arXiv arxiv.org Jul 9, 2024 2 facts
referenceKnowPhish is a phishing detection system that functions as an add-on, combining a Brand Knowledge Base (built using Wikidata) with a Large Language Model (LLM) to detect phishing attempts by extracting brand information from webpage text.
claimWikidata was utilized to create the Brand Knowledge Base for KnowPhish by providing entity information about brands, including logos, official website URLs, and aliases.
The construction and refined extraction techniques of knowledge ... nature.com Feb 10, 2026 1 fact
referenceArnaout, H. et al. published 'Negative knowledge for open-world Wikidata' in the Companion Proceedings of the Web Conference 2021, pp. 544–551 (2021).
LLM-empowered knowledge graph construction: A survey - arXiv arxiv.org Oct 23, 2025 1 fact
referenceXiaohan Feng, Xixin Wu, and Helen Meng authored 'Ontology-grounded Automatic Knowledge Graph Construction by LLM under Wikidata schema', published as an arXiv preprint in December 2024.
A survey on augmenting knowledge graphs (KGs) with large ... link.springer.com Nov 4, 2024 1 fact
referenceVrandečić and Krötzsch (2014) published 'Wikidata: a free collaborative knowledge base' in Communications of the ACM.
bureado/awesome-software-supply-chain-security - GitHub github.com 1 fact
referenceRepology provides coverage for Linux packages across multiple distributions and includes open source infrastructure like repology-updater, which also provides an updater for WikiData.
Medicinal plants meet modern biodiversity science - OUCI ouci.dntb.gov.ua 1 fact
claimThe LOTUS initiative, as described by Rutz et al., manages knowledge for open natural products research through Wikidata.