WordNet: An Electronic Lexical Database

By Dekang Lin. 1998

Subfields: Natural Language Processing, 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 Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

Grafit: Learning fine-grained image representations with coarse labels (2020, background): “To evaluate this, they consider the 1000 ImageNet classes as ﬁne, and group them into 127 coarse classes with the WordNet [20] hierarchy.”
Learning Gender-Neutral Word Embeddings (2018, methodology): “We categorize all the vocabulary words into three sub-sets: male-deﬁnition Ω M , female-deﬁnition Ω F , and gender-neutral Ω N , based on their deﬁnition in WordNet (Miller and Fellbaum, 1998).”
Movie Description (2016, methodology): “VerbNet is already mapped to Word-Net, thus we map to VerbNet via WordNet.”
SemEval-2014 Task 10: Multilingual Semantic Textual Similarity (2014, methodology): “The semantic similarity task was approached from a monolingual perspective as well (Gupta, 2014), by focusing on Spanish resources, such as the trial data we released as part of the subtask, and the Spanish WordNet; 16 these were leveraged using meta-learning over variations of overlap-based metrics”
Modeling Biological Processes for Reading Comprehension (2014, methodology): “…tag of the word and adjacent words, dependency path to the root, POS tag of children and parent in the dependency tree, and clustering features from WordNet (Fellbaum, 1998), Nomlex (Macleod et al., 1998), Levin verb classes (Levin, 1993), and a list of biological processes compiled from Wikipedia.”
Data Augmentation using Pre-trained Transformer Models (2020, background): “In particular, Wei and Zou (2019) showed that simple word replacement using knowledge bases like WordNet (Miller, 1998) improves classification performance.”
Efficient Feature Transformations for Discriminative and Generative Continual Learning (2021, methodology): “The ImageNet-1K [6] classification dataset contains 1000 classes based on the WordNet [33] hierarchy.”
Learning Word-Like Units from Joint Audio-Visual Analysis (2017, methodology): “, 2015) used to train the VGG network were derived from WordNet synsets (Fellbaum, 1998), we can measure the semantic similarity between the words learned by our network and the ILSVRC2012 class labels by using synset similarity measures within WordNet.”
Environments for Multiagent Systems State-of-the-Art and Research Challenges (2004, methodology): “Retsina provides an ontology derived from the Wordnet Ontology [27] and a protocol engine with a protocol language.”
The communicative function of ambiguity in language. (2011, unknown): “In this section we extend the analyses of the previous sections to the syllable level.”
YAGO3: A Knowledge Base from Multilingual Wikipedias (2015, background): “The online dictionary WordNet [18] aims to cover all words of the English language.”
Using Semantic Roles to Improve Question Answering (2007, background): “WordNet (Fellbaum, 1998) is perhaps the most popular resource and has been employed in a variety of QA-related tasks ranging from query expansion, to axiom-based reasoning (Moldovan et al., 2003), passage scoring (Paranjpe et al., 2003), and answer ﬁltering (Leidner et al., 2004).”
Unsupervised Ontology Induction from Text (2010, background): “A joint approach propagates in-1 formation among the three tasks, uncovers more implicit information from text, and can potentially work well even in domains not well covered by existing resources like WordNet and Wikipedia.”
Large Scale Visual Recognition (2012, methodology): “It consists of 32x32 pixel versions of images collected by performing web queries for the nouns in the WordNet [37] hierarchy, without further verification of content.”
Novel Word-sense Identification (2014, background): “…the usages of drop , wall , and blow up in the following sentences correspond to relatively-recent senses of these words that appear to be quite common in text related to popular culture, but are not listed in many dictionaries; for example, they are all missing from WordNet 3.0 (Fellbaum, 1998).”
A performance evaluation of gradient field HOG descriptor for sketch based image retrieval (2013, background): “Note that although ‘WordNet’ [63] shows the semantic structures of words and has been used for image retrieval, in our case the words relation shown by WordNet could be too broad for a particular subset of the world scenario.”
YAGO: A Large Ontology from Wikipedia and WordNet (2008, methodology): “But the existing applications typically use only a single source of background knowledge (mostly WordNet [26] or Wikipedia).”