Learning subjective nouns using extraction pattern bootstrapping

By Ellen Riloff, Janyce Wiebe, Theresa Wilson. 2003

559 citations recorded by the source corpus.

We explore the idea of creating a subjectivity classifier that uses lists of subjective nouns learned by bootstrapping algorithms. The goal of our research is to develop a system that can distinguish subjective sentences from objective sentences. First, we use two bootstrapping algorithms that exploit extraction patterns to learn sets of subjective nouns. Then we train a Naive Bayes classifier using the subjective nouns, discourse features, and subjectivity clues identified in prior research. The bootstrapping algorithms learned over 1000 subjective nouns, and the subjectivity classifier performed well, achieving 77% recall with 81% precision.

Subfields: Artificial Intelligence

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

Learning Extraction Patterns for Subjective Expressions (2003, methodology): “One classiﬁer searches the unannotated corpus for sentences that can be labeled as subjective with high conﬁdence, and the other classiﬁer searches for sentences that can be labeled as objective with high conﬁdence.”
Twitter mood predicts the stock market (2010, methodology): “We adopt OF’s subjective lexicon that has been established upon previous work [37], [38], [24].”
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).”
SENTIWORDNET: A Publicly Available Lexical Resource for Opinion Mining (2006, background): “The task of determining whether a term is indeed a marker of opinionated content (i.e. is Subjective or Objective) has instead received much less attention (Esuli and Sebastiani, 2006; Riloff et al., 2003; Vegnaduzzo, 2004).”
A machine learning approach to sentiment analysis in multilingual Web texts (2009, methodology): “An opinion word lexicon was constructed by Riloff et al. (2003) , while Hu and Liu (2004) identify words describing the features of a product (e.g., “The camera takes incredible pictures.”) and use them in the classification.”
Sentiment Analysis: An Overview from Linguistics (2016, unknown): “The syntactic clues are learned from manually annotated data (Riloff et al 2003, Wiebe et al 2003).”
Determining the Sentiment of Opinions (2004, background): “Recent computational work either focuses on sentence ‘subjectivity’ (Wiebe et al. 2002; Riloff et al. 2003), concentrates just on explicit statements of evaluation, such as of films (Turney 2002; Pang et al. 2002), or focuses on just one aspect of opinion, e.g., (Hatzivassiloglou and McKeown 1997)…”
Semi-Supervised Polarity Lexicon Induction (2009, methodology): “In a different work, Riloff et al. (2003) use manually derived pattern templates to extract subjective nouns by bootstrapping.”
Which Side are You on? Identifying Perspectives at the Document and Sentence Levels (2006, background): “, 2003), identifying opinionated documents (Yu and Hatzivassiloglou, 2003) and sentences (Yu and Hatzivassiloglou, 2003; Riloff et al., 2003), and discriminating between positive and negative language (Pang et al.”
Mining User Intents in Twitter: A Semi-Supervised Approach to Inferring Intent Categories for Tweets (2015, methodology): “We adopt the bootstrapping based method (Riloff, Wiebe, and Wilson 2003; Zhao et al. 2014) to retrieve intent tweets.”
Bootstrapping Without the Boot (2005, methodology): “Riloff et al. (2003) learned lists of subjective nouns in English, seeding their method with 20 high-frequency, strongly subjective words.”
Sentiment Analysis and Subjectivity (2010, methodology): “To save the manual labeling effort, a bootstrapping approach to label training data automatically is reported in [80, 81].”
Learning Subjective Language (2004, background): “The most closely related new work is Riloff, Wiebe, and Wilson (2003), Riloff and Wiebe (2003) and Yu and Hatzivassiloglou (2003).”
Aspect-based sentiment analysis of movie reviews on discussion boards (2010, methodology): “The words were extracted and expanded from General Inquirer and the work of Hatzivassiloglou and McKeown [10] and Riloff et al. [12].”
Towards a Unified Natural Language Inference Framework to Evaluate Sentence Representations (2018, background): “, 2006), as opposed to differentiating between subjective and objective information (Yu and Hatzivassiloglou, 2003; Riloff et al., 2003).”