DBpedia - A large-scale, multilingual knowledge base extracted from Wikipedia

By Jens Lehmann, Robert Isele, Max Jakob, Anja Jentzsch, Dimitris Kontokostas, Pablo N. Mendes, Sebastian Hellmann, Mohamed Morsey, Patrick van Kleef, Sören Auer and 1 more. 2015

3,616 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Communication, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Communication

In Evaluation & Benchmarks

Citations in the corpus

Semi-supervised Sequence Learning (2015, methodology): “A DBpedia document has an average of 300 characters while the maximum length of all documents is 13,467 characters.”
Generating Wikipedia by Summarizing Long Sequences (2018, background): “There is a rich body of work incorporating Wikipedia for machine learning tasks, including question-answering (Hewlett et al. (2016), Rajpurkar et al. (2016)) and information extraction (Lehmann et al., 2015), and text generation from structured data (Lebret et al., 2016).”
Snorkel: Rapid Training Data Creation with Weak Supervision (2017, methodology): “DBpedia -”
Word-Entity Duet Representations for Document Ranking (2017, background): “Recently, large general domain knowledge bases, such as DBpe-dia [11] and Freebase [1], have emerged.”
Character-level Convolutional Networks for Text Classification (2015, methodology): “82 39.30 28.80 40.45 4.93 Sm. Conv. Th. 14.80 - 1.85 6.49 40.16 29.84 40.43 5.67 DBPedia ontology dataset. DBpedia is a crowd-sourced community effort to extract structured information from Wikipedia [19]. The DBpedia ontology dataset is constructed by picking 14 nonoverlapping classes from DBpedia ”
Symbolic Knowledge Distillation: from General Language Models to Commonsense Models (2021, methodology): “Past automatic data mainly use extractive approaches, e.g. syntactic parsing (Zhang et al., 2020a) or pattern matching (Li et al., 2020) from unstructured text (Lehmann et al., 2015; Buck et al., 2014).”
Zero-shot Text Classification With Generative Language Models (2019, methodology): “We evaluate the trained model on the DBPedia, AGNews, Yahoo Answers, SST-2, Amazon-2, and Yelp-2 text classiﬁcation datasets (Socher et al., 2013; Lehmann et al., 2015).”
EDUCE: Explaining model Decisions through Unsupervised Concepts Extraction (2019, methodology): “The DBpedia ontology classiﬁcation dataset was constructed by picking 14 non-overlapping categories from DBpedia 2014 [14].”
LFPT5: A Unified Framework for Lifelong Few-shot Language Learning Based on Prompt Tuning of T5 (2021, neutral): “…experiments on four different datasets/domains: AGNews for news classiﬁcation (Zhang et al., 2015), Amazon Review for sentiment analysis (McAuley et al., 2015), DBPedia for Wikipedia article classiﬁcation into topics (Lehmann et al., 2015), and Yahoo for QA categorization (Zhang et al., 2015).”
Open Question Answering with Weakly Supervised Embedding Models (2014, background): “The emergence of large-scale KBs, such as Freebase [4] or DBpedia [15], changed the setting by transforming open question answering into a problem of querying a KB using natural language.”
Visual question answering: Datasets, algorithms, and future challenges (2016, methodology): “The external knowledge bases were tailored to general information obtained from DBpedia [67], so it is possible that using a source tailored to VQA could yield greater improvement.”
Don't just listen, use your imagination: Leveraging visual common sense for non-visual tasks (2015, background): “Most existing common sense knowledge bases rely on knowledge described via text – either mined [6, 24, 29] or manually entered [33, 39, 5, 40].”
StarSpace: Embed All The Things! (2017, background): “• DBpedia (Lehmann et al. 2015) is a 14 class classification problem given the title and abstract of Wikipedia articles as input.”
Women through the glass ceiling: gender asymmetries in Wikipedia (2016, background): “DBpedia [22] is a structured version of Wikipedia that provides meta-data for articles; normalized article Uniform Resource Identifiers (URIs) that allow to interlink articles about the same entity in different language editions; normalized links between articles (taking care of redirections); and a”
Deep Text Classification Can be Fooled (2017, methodology): “ A recent work (Zhang et al. 2015) presents a convolutional network model for text classification at character level. For model training and evaluation, the authors employ a DBpedia ontology dataset (Lehmann et al. 2014), which contains 560,000 training samples and 70,000 test samples of 14 high-lev”