Unsupervised Large-Vocabulary Word Sense Disambiguation with Graph-based Algorithms for Sequence Data Labeling
By Rada Mihalcea. Human Language Technology - The Baltic Perspectiv, 2005
300 citations recorded by the source corpus.
This paper introduces a graph-based algorithm for sequence data labeling, using random walks on graphs encoding label dependencies. The algorithm is illustrated and tested in the context of an unsupervised word sense disambiguation problem, and shown to significantly outperform the accuracy achieved through individual label assignment, as measured on standard sense-annotated data sets.
Subfields: Artificial Intelligence, Information Systems
View Paper: https://www.semanticscholar.org/paper/7dbbf66450ab4ddc2bf80c65d4913b7c6fffb75e
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
- Roberto Navigli
- Mirella Lapata
- Alessandro Raganato
- Mohammad Taher Pilehvar
- Percy Liang
- Alessandro Moschitti
- Jianfeng Gao
- Jordan L. Boyd-Graber
In Information Systems
- Roberto Navigli
- Mirella Lapata
- Alessandro Raganato
- Andrea Moro
- Jordan L. Boyd-Graber
- Roberto Basili
- Alessandro Moschitti
- Eneko Agirre
Citations in the corpus
- Personalizing PageRank for Word Sense Disambiguation (2009, methodology): “In this section we present the application of PageR-ank to WSD.”
- Structured Lexical Similarity via Convolution Kernels on Dependency Trees (2011, methodology): “graphs, are enough similar, as explored in (Mihalcea, 2005; Zhao et al., 2009; Fürstenau and Lapata, 2009; Navigli and Lapata, 2010).”
- BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network (2012, methodology): “We follow previous work [76,73] and apply PageRank to induce the ranking of the senses.”
- A Topic Model for Word Sense Disambiguation (2007, methodology): “WORDNET, a lexical resource designed by psychologists and lexicographers to mimic the semantic organization in the human mind, links “synsets” (short for synonym sets) with myriad connections.”
- An Experimental Study of Graph Connectivity for Unsupervised Word Sense Disambiguation (2010, methodology): “PageRank assigns a numerical weighting to each element of a hyperlinked set of documents, with the purpose of measuring its relative importance within the set, whereas HITS rates Web pages for their authority and hub values.”
- Entity Linking meets Word Sense Disambiguation: a Unified Approach (2014, background): “Some of these approaches marginally take into account the structural properties of the knowledge base (Mihalcea, 2005).”