Representing General Relational Knowledge in ConceptNet 5
By Robyn Speer, Catherine Havasi. 2012
515 citations recorded by the source corpus.
ConceptNet is a knowledge representation project, providing a large semantic graph that describes general human knowledge and how it is expressed in natural language. This paper presents the latest iteration, ConceptNet 5, including its fundamental design decisions, ways to use it, and evaluations of its coverage and accuracy.
Subfields: Artificial Intelligence, Computational Theory and Mathematics, Knowledge Graphs & Representation
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
- Mohit Bansal
- Yejin Choi
- Luke S. Zettlemoyer
- Hannaneh Hajishirzi
- Jianfeng Gao
- Noah A. Smith
- Percy Liang
- Hinrich Schütze
In Computational Theory and Mathematics
- Lisa Bauer
- Xiang Ren
- Tom'avs Kovcisk'y
- Gautier Izacard
- Alexander H. Miller
- Dawn Song
- Mohit Bansal
- Wen-tau Yih
In Knowledge Graphs & Representation
- Xiang Ren
- Yejin Choi
- Ronan Le Bras
- Minlie Huang
- Mohit Bansal
- Chandra Bhagavatula
- Peter West
- Bill Yuchen Lin
Citations in the corpus
- Language Models as Knowledge Bases? (2019, background): “ConceptNet (Speer and Havasi, 2012) is a multi-lingual knowledge base, initially built on top of Open Mind Common Sense (OMCS) sentences.”
- The Web Is Your Oyster - Knowledge-Intensive NLP against a Very Large Web Corpus (2021, background): “There have been many attempts to capture such implicit knowledge vie common sense knowledge bases (Speer and Havasi, 2012; Sap et al., 2019a; Bhakthavatsalam et al., 2020).”
- Factual Probing Is [MASK]: Learning vs. Learning to Recall (2021, background): “LAMA facts are drawn from a number of sources, including Wikidata, ConceptNet (Speer and Havasi, 2012), and SQuAD (Rajpurkar et al., 2016).”
- ERNIE: Enhanced Language Representation with Informative Entities (2019, background): “There are three important directions remain for future research: (1) inject knowledge into feature-based pre-training models such as ELMo (Peters et al., 2018); (2) introduce diverse structured knowledge into language representation models such as Con-ceptNet (Speer and Havasi, 2012) which is differ”
- Commonsense Knowledge Mining from Pretrained Models (2019, methodology): “Using the approach described by Speer and Havasi (2012), and followed by Li et al. (2016) and Jastrz˛ebski et al. (2018), two human annotators manually rate the 100 triples with the highest predicted score on a 0 to 4 scale: 0 (Doesn’t make sense), 1 (Not true), 2 (Opinion/Don’t know), 3 (Sometimes…”
- Unsupervised Commonsense Question Answering with Self-Talk (2020, methodology): “To increase the coverage of high-precision world knowledge and facilitate multi-hop reasoning by making intermediate reasoning steps explicit, prior work incorporated KBs (e.g. ConceptNet; Speer and Havasi, 2012) and knowledge-informed models into LM-based models (Xia et al.”
- Commonsense Knowledge Base Completion (2016, methodology): “To address this, we include an additional nonlinear transformation of each term: 2 These examples are from the Open Mind Common Sense (OMCS) part of ConceptNet version 5 (Speer and Havasi, 2012).”
- When Do You Need Billions of Words of Pretraining Data? (2020, background): “, 2018), ConceptNet (Speer and Havasi, 2012), and SQUAD (Rajpurkar et al.”
- Getting Closer to AI Complete Question Answering: A Set of Prerequisite Real Tasks (2020, methodology): “The TriAN architecture incorporates knowledge from the ConceptNet knowledge base (Speer and Havasi 2012).”
- Commonsense Knowledge Aware Conversation Generation with Graph Attention (2018, methodology): “…systems, common-sense knowledge is important for establishing effective interactions, since socially shared commonsense knowledge is the set of background information people intended to know and use during conversation [Minsky, 1991; Markov´a et al. , 2007; Speer and Havasi, 2012; Souto, 2015].”
- AMMUS : A Survey of Transformer-based Pretrained Models in Natural Language Processing (2021, background): “Negated LAMA [253] Impact of negation in probing factual and common-sense knowledge English UnTQ Single token entities TRex [249], GoogleRE, ConceptNet [250]and SQUAD [251]”
- Commonsense for Generative Multi-Hop Question Answering Tasks (2018, methodology): “Recently, largescale graphical commonsense databases such as ConceptNet (Speer and Havasi, 2012) use graphical structure to express intricate relations between concepts, but effective goal-oriented graph traversal has not been extensively used in previous commonsense incorporation efforts.”
- Reading Twice for Natural Language Understanding (2017, background): “ConceptNet ConceptNet1 [23] (CN) is freely-available, multi-lingual semantic network that originated from the Open Mind Common Sense project and incorporates selected knowledge from various other knowledge sources, such as Wiktionary2, Open Multilingual WordNet3, OpenCyc or DBpedia4.”
- UNION: An Unreferenced Metric for Evaluating Open-ended Story Generation (2020, methodology): “We use the commonsense knowledge base Con-ceptNet (Speer and Havasi, 2012) 3 for keyword recognition and antonym query.”
- AliCoCo: Alibaba E-commerce Cognitive Concept Net (2020, background): “ConceptNet [27] tries to include common sense knowledge by recognizing informal relations between concepts, where the concepts could be the conceptualization of any human knowledge such as “games with a purpose” appearing in free texts.”
- Distilling the Outcomes of Personal Experiences: A Propensity-scored Analysis of Social Media (2017, methodology): “While the first version mainly relied on data from the Open Mind Common Sense project [90], now ConceptNet5’s sources range “from dictionaries to online games” [91], including DBPedia [7], English Wiktionary15, and WordNet 3.”
- Do Neural Language Models Overcome Reporting Bias? (2020, background): “Apart from several notable efforts to collect commonsense knowledge from experts (Lenat, 1995) or through crowdsourcing (Speer and Havasi, 2012; Sap et al., 2019), most work has been on extracting such knowledge from large text corpora (Mitchell et al., 2018).”