A Bootstrapping Method for Learning Semantic Lexicons using Extraction Pattern Contexts

By Michael Thelen, Ellen Riloff. Conference on Empirical Methods in Natural Language Processing, 2002

426 citations recorded by the source corpus.

This paper describes a bootstrapping algorithm called Basilisk that learns high-quality semantic lexicons for multiple categories. Basilisk begins with an unannotated corpus and seed words for each semantic category, which are then bootstrapped to learn new words for each category. Basilisk hypothesizes the semantic class of a word based on collective information over a large body of extraction pattern contexts. We evaluate Basilisk on six semantic categories. The semantic lexicons produced by Basilisk have higher precision than those produced by previous techniques, with several categories showing substantial improvement.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/915ceac70544cf69202b74cc4582e4d42ab7ed46

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

Citations in the corpus

Detecting offensive tweets via topical feature discovery over a large scale twitter corpus (2012, methodology): “Research that applied bootstrapping to the NLP domain existed more than a decade ago (Jones et al., 1999; Thelen and Riloff, 2002).”
Improved Pattern Learning for Bootstrapped Entity Extraction (2014, methodology): “Thelen and Riloff (2002) extended the above bootstrapping algorithm for multi-class learning.”
A survey on sentiment detection of reviews (2009, neutral): “…1998; Grefenstette, 1994; Justeson & Slava, 1993; Kamps & Maarten, 2002; Rapp, 2004; Riloff & Shepherd, 1997; Roark & Charniak, 1998; Taboada, 2006; Thelen & Riloff, 2002; Valitutti, Strapparava, & Stock, 2004), the other based on part-of-speech (POS) tags (Ait-Mokhtar, Chanod, & Roux, 2002;…”
Exploiting Wikipedia as External Knowledge for Named Entity Recognition (2007, methodology): “Many methods have been proposed for solving this problem by automatically extracting gazetteers from large amounts of texts (Riloff and Jones, 1999; Thelen and Riloff, 2002; Etzioni et al., 2005; Shinzato et al., 2006; Talukdar et al., 2006; Nadeau et al., 2006).”
Fine-grained subjectivity and sentiment analysis: recognizing the intensity, polarity, and attitudes of private states (2008, methodology): “” In (Riloff, Wiebe, and Wilson, 2003), the Meta-Bootstrapping (Riloff and Jones, 1999) and Basilisk (Thelen and Riloff, 2002) bootstrapping algorithms were used to learn sets of subjective nouns.”
Bootstrapped Learning of Semantic Classes from Positive and Negative Examples (2003, neutral): “• As observed by Thelen and Riloff (2002), a category is less likely to expand beyond its true territory if it cannot acquire names that have already been claimed by other categories.”
Structured learning with constrained conditional models (2012, background): “Therefore, an increased attention has been recently given to semi-supervised learning, where large amounts of unlabeled data are used to improve models learned from a small training set (Yarowsky 1995; Blum and Mitchell 1998; Collins and Singer 1999; Thelen and Riloff 2002; Haghighi and Klein 2006).”
Incorporating Lexical Priors into Topic Models (2012, background): “For each document d , choose θ d ∼ Dir ( α ) .”
Semi-automatic construction of a named entity dictionary for entity-based sentiment analysis in social media (2017, methodology): “These methods mainly use linguistic patterns based on statistical information [1,15,17].”
Classifying Sentences as Speech Acts in Message Board Posts (2011, methodology): “Semantic Lexicon: Basilisk (Thelen and Riloff, 2002) is a bootstrapping algorithm that has been used to induce semantic lexicons for terrorist events (Thelen and Riloff, 2002), biomedical concepts (McIntosh, 2010), and subjective/objective nouns for opinion analysis (Riloff et al., 2003).”
Learning subjective nouns using extraction pattern bootstrapping (2003, methodology): “In the last few years, two bootstrapping algorithms have been developed to create semantic dictionaries by exploiting extraction patterns: Meta-Bootstrapping (Riloff and Jones, 1999) and Basilisk (Thelen and Riloff, 2002).”
Inducing Domain-Specific Sentiment Lexicons from Unlabeled Corpora (2016, background): “These patterns may rely on syntactic structures (Hatzivassiloglou and McKeown, 1997; Jijkoun et al., 2010; Rooth et al., 1999; Thelen and Riloff, 2002; Widdows and Dorow, 2002), which can be domain-specific and brittle (e.”
Guiding Semi-Supervision with Constraint-Driven Learning (2007, methodology): “This was used, for example, by (Thelen and Riloff, 2002; Collins and Singer, 1999) in information extraction, and by (Smith and Eisner, 2005) in POS tagging.”