Learning Extraction Patterns for Subjective Expressions

  title={Learning Extraction Patterns for Subjective Expressions},
  author={Ellen Riloff and Janyce Wiebe},
  booktitle={Conference on Empirical Methods in Natural Language Processing},
  • E. RiloffJ. Wiebe
  • Published in
    Conference on Empirical…
    11 July 2003
  • Computer Science
This paper presents a bootstrapping process that learns linguistically rich extraction patterns for subjective (opinionated) expressions. High-precision classifiers label unannotated data to automatically create a large training set, which is then given to an extraction pattern learning algorithm. The learned patterns are then used to identify more subjective sentences. The bootstrapping process learns many subjective patterns and increases recall while maintaining high precision. 

Figures and Tables from this paper

Automatic Detection of Subjective Sentences Based on Chinese Subjective Patterns

A bootstrapping process that can use subjective phrases to automatically create training set from unannotated data, which is then fed to a subjective phrases extraction algorithm, which learns phrases that are used to identify more subjective sentences.

Creating Subjective and Objective Sentence Classifiers from Unannotated Texts

The results of developing subjectivity classifiers using only unannotated texts for training rivals that of previous supervised learning approaches and advances the state of the art in objective sentence classification.


A bootstrapping technique for identifying paraphrases for emotion terms, from nonparallel corpora, using annotated blogs and other data sets as texts as texts from which to extract paraphrasing, based on the highest scoring extraction patterns.

Identifying Sources of Opinions with Conditional Random Fields and Extraction Patterns

This work adopts a hybrid approach that combines Conditional Random Fields (Lafferty et al., 2001) and a variation of AutoSlog (Riloff, 1996a), and shows that the combination of these two methods performs better than either one alone.

Learning Opinionated Patterns for Contextual Opinion Detection

This paper proposes and evaluates an original hybrid method, based on syntactic information extraction and clustering techniques, to learn automatically such patterns and integrate them into an opinion detection system.

Building a Sentiment Lexicon for Social Judgement Mining

A methodology for automatically enlarging a Portuguese sentiment lexicon for mining social judgments from text, i.e., detecting opinions on human entities, produces results at least as good as the best that have been reported for this task.

Automatic extraction and classification approach of opinions in texts

A similarity measurement calculating semantically distances between a word and predefined subgroups of seed words is proposed and obtained the best value of Precision and F1 62% and 61%.

Exploiting Subjectivity Classification to Improve Information Extraction

An IE system that uses a subjective sentence classifier to filter its extractions is described and it is found that indiscriminately filtering extractions from subjective sentences was overly aggressive, but more selective filtering strategies improved IE precision with minimal recall loss.

Integrating Knowledge for Subjectivity Sense Labeling

This paper introduces an integrative approach to automatic word sense subjectivity annotation. We use features that exploit the hierarchical structure and domain information in lexical resources such

Extracting opinion sentence by combination of SVM and syntactic templates

  • Bo ZhangYanquan ZhouYu Mao
  • Computer Science
    Proceedings of the 6th International Conference on Natural Language Processing and Knowledge Engineering(NLPKE-2010)
  • 2010
A combined method of syntactic structure, dependency relation and SVM classifier to extract opinion sentences that performed well, achieving 92.6% recall with 85.5% precision.



Learning subjective nouns using extraction pattern bootstrapping

The goal of the research is to develop a system that can distinguish subjective sentences from objective sentences, and a Naive Bayes classifier is trained using the subjective nouns, discourse features, and subjectivity clues identified in prior research.

Learning Subjective Adjectives from Corpora

This paper identifies strong clues of subjectivity using the results of a method for clustering words according to distributional similarity (Lin 1998), seeded by a small amount of detailed manual annotation.

Learning information extraction patterns from examples

  • S. Huffman
  • Computer Science
    Learning for Natural Language Processing
  • 1995
A system that can learn dictionaries of extraction patterns directly from user-provided examples of texts and events to be extracted from them, and learns patterns that recognize relationships between key constituents based on local syntax.

Identifying Collocations for Recognizing Opinions

Promising results are shown for a straightforward method of identifying collocational clues of subjectivity, as well as evidence of the usefulness of these clues for recognizing opinionated documents.

Learning Dictionaries for Information Extraction by Multi-Level Bootstrapping

A multilevel bootstrapping algorithm is presented that generates both the semantic lexicon and extraction patterns simultaneously simultaneously and produces high-quality dictionaries for several semantic categories.

Relational learning techniques for natural language information extraction

A novel rule representation speci c to natural language and a learning system, Rapier, which learns information extraction rules, and initial results on a small corpus of computer-related job postings with a preliminary version of Rapier are presented.

Acquisition of semantic patterns for information extraction from corpora

  • J.-T. KimD. Moldovan
  • Computer Science
    Proceedings of 9th IEEE Conference on Artificial Intelligence for Applications
  • 1993
A knowledge acquisition tool to extract semantic patterns for a memory-based information retrieval system is presented to facilitate the construction of a large knowledge base of semantic patterns.

Learning Information Extraction Rules for Semi-Structured and Free Text

WHISK is designed to handle text styles ranging from highly structured to free text, including text that is neither rigidly formatted nor composed of grammatical sentences, and can also handle extraction from free text such as news stories.

CRYSTAL: Inducing a Conceptual Dictionary

CRYSTAL is described, a system which automatically induces a dictionary of "concept-node definitions" sufficient to identify relevant information from a training corpus that can often surpass human intuitions in creating reliable extraction rules.

Automatic Acquisition of Domain Knowledge for Information Extraction

In developing an Information Extraction (IE) system for a new class of events or relations, one of the major tasks is identifying the many ways in which these events or relations may be expressed in