Thumbs up?
Published 1 January 2002Open access
Bo Pang, Lillian Lee, Shivakumar Vaithyanathan
Citations6,987
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This work considers the problem of classifying documents not by topic, but by overall sentiment, e.g., determining whether a review is positive or negative, and concludes by examining factors that make the sentiment classification problem more challenging.
Abstract
We consider the problem of classifying documents not by topic, but by overall sentiment, e.g., determining whether a review is positive or negative. Using movie reviews as data, we find that standard machine learning techniques definitively outperform human-produced baselines. However, the three machine learning methods we employed (Naive Bayes, maximum entropy classification, and support vector machines) do not perform as well on sentiment classification as on traditional topic-based categorization. We conclude by examining factors that make the sentiment classification problem more challenging.
Keywords
Computer Science
Lecture notes in computer scienceText categorization with Support Vector Machines: Learning with many relevant features
7,925 Citations1998Thorsten Joachims
SVMs achieve substantial improvements over the currently best performing methods and behave robustly over a variety of di-erent learning tasks, eliminating the need for manual parameter tuning.
Cambridge University Press eBooksVariation across Speech and Writing
5,128 Citations1988Douglas Biber
In Variation Across Speech and Writing, six dimensions of variation are identified through a factor analysis, on the basis of linguistic co-occurence patterns, and the resulting model of variation provides for the description of the distinctive linguistic characteristic of any spoken or written text and enables reconciliation of the contradictory conclusions reached in previous research.
Technical reportsMaking Large-Scale SVM Learning Practical
4,317 Citations2006Thorsten Joachims
This chapter presents algorithmic and computational results developed for SVM light V 2.0, which make large-scale SVM training more practical and give guidelines for the application of SVMs to large domains.
Meeting of the Association for Computational LinguisticsThumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews
3,654 Citations2002Peter Peter, Turney
A comparison of event models for naive bayes text classification
3,224 Citations1998Andrew McCallum, Kamal Nigam
It is found that the multi-variate Bernoulli performs well with small vocabulary sizes, but that the multinomial performs usually performs even better at larger vocabulary sizes--providing on average a 27% reduction in error over the multi -variateBernoulli model at any vocabulary size.
Computational LinguisticsA maximum entropy approach to natural language processing
3,120 Citations1996Adam Berger, Vincent J. Della Pietra +1 more
A maximum-likelihood approach for automatically constructing maximum entropy models is presented and how to implement this approach efficiently is described, using as examples several problems in natural language processing.
Machine LearningOn the Optimality of the Simple Bayesian Classifier under Zero-One Loss
3,072 Citations1997Pedro Domingos, Michael J. Pazzani
The Bayesian classifier is shown to be optimal for learning conjunctions and disjunctions, even though they violate the independence assumption, and will often outperform more powerful classifiers for common training set sizes and numbers of attributes, even if its bias is a priori much less appropriate to the domain.
Lecture notes in computer scienceNaive (Bayes) at forty: The independence assumption in information retrieval
2,092 Citations1998David Lewis
The naive Bayes classifier, currently experiencing a renaissance in machine learning, has long been a core technique in information retrieval, and some of the variations used for text retrieval and classification are reviewed.
ArXiv.orgThumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews
1,585 Citations2002Peter D. Turney
Predicting the semantic orientation of adjectives
1,439 Citations1997Vasileios Hatzivassiloglou, Kathleen McKeown
A log-linear regression model uses constraints from conjunctions to predict whether conjoined adjectives are of same or different orientations, achieving 82% accuracy in this task when each conjunction is considered independently.
IEEE Transactions on Pattern Analysis and Machine IntelligenceInducing features of random fields
1,044 Citations1997S. Della Pietra, V. Della Pietra +1 more
The random field models and techniques introduced in this paper differ from those common to much of the computer vision literature in that the underlying random fields are non-Markovian and have a large number of parameters that must be estimated.
Predicting the semantic orientation of adjectives
819 Citations1997Vasileios Hatzivassiloglou, Kathleen McKeown
Using Maximum Entropy for Text Classification
756 Citations1999Kamal Nigam, John Lafferty +1 more
This paper uses maximum entropy techniques for text classification by estimating the conditional distribution of the class variable given the document by comparing accuracy to naive Bayes and showing that maximum entropy is sometimes significantly better, but also sometimes worse.
Effects of adjective orientation and gradability on sentence subjectivity
671 Citations2000Vasileios Hatzivassiloglou, Janyce Wiebe
A novel trainable method that statistically combines two indicators of gradability is presented and evaluated, complementing existing automatic techniques for assigning orientation labels.
Communications of the ACMPHOAKS
496 Citations1997Loren Terveen, Will Hill +3 more
PHOAKS is distinguished from other recommender systems by two major design principles: role specialization and reuse and the number of distinct recommenders of a resource is a plausible measure of resource quality.
arXiv (Cornell University)Unsupervised Learning of Semantic Orientation from a Hundred-Billion-Word Corpus
361 Citations2002Peter D. Turney, Michael L. Littman
This paper introduces a simple algorithm for unsupervised learning of semantic orientation from extremely large corpora by issuing queries to a Web search engine and using pointwise mutual information to analyse the results.
Automatic detection of text genre
358 Citations1997Brett Kessler, Geoffrey Numberg +1 more
A theory of genres as bundles of facets, which correlate with various surface cues, are proposed, and it is argued that genre detection based on surface cues is as successful as Detection based on deeper structural properties.
Recognizing text genres with simple metrics using discriminant analysis
293 Citations1994Jussi Karlgren, Douglass R. Cutting
A simple method for categorizing texts into pre-determined text genre categories using the statistical standard technique of discriminant analysis is demonstrated with application to the Brown corpus.
arXiv (Cornell University)Automatic Detection of Text Genre
206 Citations1997Brett Kessler, Geoffrey Nunberg +1 more
IEEE Transactions on Speech and Audio ProcessingA survey of smoothing techniques for ME models
199 Citations2000S.F. Chen, Roni Rosenfeld
Over a large number of data sets, it is found that fuzzy ME smoothing performs as well as or better than all other algorithms under consideration.
Smokey: automatic recognition of hostile messages
170 Citations1997Ellen Spertus
Some approaches to flame recognition are described, mcluding a prototype system, Smokey, which builds a 47-element feature vector based on the syntax and semantics of each sentence, combining the vectors for the sentences within each message.
Identifying Collocations for Recognizing Opinions
135 Citations2001Jens Wiebe
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.
Direction-based text interpretation as an information access refinement
135 Citations1992Marti A. Hearst
A method is described for coercing sentence meanings into a metaphoric model such that the only semantic interpretation needed in order to determine the directionality of a sentence is done with respect to the model.
A decision tree of bigrams is an accurate predictor of word sense
129 Citations2001Ted Pedersen
This paper presents a corpus-based approach to word sense disambiguation where a decision tree assigns a sense to an ambiguous word based on the bigrams that occur nearby.
Natural Language EngineeringThe grammar of sense: Using part-of-speech tags as a first step in semantic disambiguation
87 Citations1998Yorick Wilks, Mark Stevenson
This paper describes two experiments: one exploring the amount of information relevant to sense disambiguation contained in the part-of-speech field of entries in a Machine Readable Dictionary (MRD); the other, more practical, experiment attempts sense disAmbiguation of all content words in a text assigning MRD homographs as sense tags using only part- of-speech information.
Lecture notes in computer scienceGenre Classification and Domain Transfer for Information Filtering
76 Citations2002Aidan Finn, Nicholas Kushmerick +1 more
This work investigates three approaches to automatically classifying documents by genre: traditional bag of words techniques, part-of-speech statistics, and hand-crafted shallow linguistic features, particularly in domain transfer conditions.
Fuzzy Typing for Document Management
52 Citations2000Alison K. Huettner
This prototype system demonstrates a novel method of document analysis and management, based on a combination of techniques from NLP and fuzzy logic, which is referred to as semantic typing.
Virtual reviewers for collaborative exploration of movie reviews
50 Citations2000Junichi Tatemura
A collaborative exploration system that helps users to explore recommendations from various viewpoints by providing “virtual reviewers” that represent particular viewpoints.
National Conference on Artificial IntelligenceOn the computation of point of view
44 Citations1994Warren Sack
The focus of the current research is to build a less gullible story understander by encoding in it a means to recognize point of view, not only for information retrieval tasks which demand a search for credible stories, but also in future entertainment technologies which will be capable of finding and then assembling together into a unified presentation.