---
title: "Effects of Adjective Orientation and Gradability on Sentence Subjectivity (2000)"
description: "For gradability, they introduce a new trainable log-linear model that combines two corpus indicators. 82% accuracy. 19 references, cited by 75 papers here."
authors: ["Vasileios Hatzivassiloglou", "Janyce M. Wiebe"]
year: 2000
venue: "COLING 2000 Volume 1: The 18th International Conference on Computational Linguistics"
source: "https://aclanthology.org/C00-1044/"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/hatzivassiloglou-2000-effects-adjective-orientation-gradability-sentence/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/hatzivassiloglou-2000-effects-adjective-orientation-gradability-sentence/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2000](https://md.research.elamur.ai/free/en/papers/2000/) / Effects of Adjective Orientation and Gradability on Sentence Subjectivity

# Effects of Adjective Orientation and Gradability on Sentence Subjectivity

**Authors:** [Vasileios Hatzivassiloglou](https://md.research.elamur.ai/free/en/authors/vasileios-hatzivassiloglou/) and [Janyce M. Wiebe](https://md.research.elamur.ai/free/en/authors/janyce-wiebe/) · **Year:** 2000 · **Venue:** [COLING 2000 Volume 1: The 18th International Conference on Computational Linguistics](https://md.research.elamur.ai/free/en/venues/coling/) ([COLING 2000](https://md.research.elamur.ai/free/en/venues/coling/2000/)) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/C00-1044/) · [PDF](https://aclanthology.org/C00-1044.pdf) · [OpenAlex](https://openalex.org/W1998442272)

**Citations:** 671 citations (OpenAlex, 2026-10-14) · [75](https://md.research.elamur.ai/free/en/papers/hatzivassiloglou-2000-effects-adjective-orientation-gradability-sentence/cited-by/) in our library, of the 141,840 papers whose bibliographies we read

## Summary
The paper investigates whether particular lexical properties of adjectives can signal that a sentence expresses opinion rather than fact. Earlier work had shown that the mere presence of adjectives correlates with subjectivity. The authors ask whether three finer-grained adjective classes sharpen this signal: dynamic adjectives, semantically oriented (positive/negative) adjectives, and gradable adjectives.

For orientation, they reuse the conjunction-based method of Hatzivassiloglou and McKeown (1997), trained on Wall Street Journal text. For gradability, they introduce a new trainable log-linear model that combines two corpus indicators. The first is how often an adjective appears with grading modifiers. The second is how often it appears in comparative or superlative form.

Each adjective class, whether labeled manually or automatically, is then tested as a predictor of subjective sentences. All three classes turn out to be strong cues, and intersecting the features often raises predictability further. Notably, the automatically derived sets perform as well as or better than the hand-labeled ones. This suggests that useful subjectivity clues can be learned from corpora for a new genre.

## Key points
- The orientation resource covers 1,336 hand-labeled oriented adjectives (657 positive, 679 negative) from a 21-million-word WSJ corpus. Independent judges agreed with the selection 89% of the time and with the labels 97% of the time.
- The conjunction-based model predicts same/different orientation links with 82% accuracy. Automatic labels cover 730 adjectives with 81.51% accuracy, using the label from the highest applicable conjunction-count threshold.
- The gradability indicators are counts of modification by a hand-compiled list of 73 grading adverbs and phrases (e.g., *very*, *somewhat*), plus counts of degree inflection detected by a custom morphology component. Text was tagged with Church's PARTS tagger.
- Dynamic adjectives reach a predictability of 74%. Automatically identified negative-polarity and gradable sets outperform their manual counterparts by about 7 and 5 percentage points respectively. Non-gradable adjectives are weak predictors (about 0.54–0.55).
- Combining features, for example gradable plus polar or dynamic within those sets, generally maintains or improves predictability.

## Interesting ideas
- **Choose the most reliable experiment per item.** A word gets labels from several runs with increasing minimum-evidence thresholds, and the label from the strictest applicable run is kept. This is a simple way to merge results whenever confidence grows with data density.
- **Turn counts into a 2×2 table and feed it to a statistical model.** Gradable contexts are weighed against non-gradable ones, so the model is not fooled by occasional exceptional uses such as *very dead*. The same idea works for any lexical property signaled by noisy contextual markers.
- **Generate-and-verify morphology.** Candidate base forms are produced by stripping suffixes (e.g., *bigger* → *big/bigg/bigge*), and the corpus frequency of each candidate picks the real one. The original word stays among the candidates to protect forms like *silver*. This is a cheap, lexicon-free trick for languages with limited resources.
- **Exploit discourse constraints as weak supervision.** Conjunctions such as "and" and "but" imply same or opposite polarity, which yields a word graph that can be clustered into polarity classes. This could transfer to other languages that have comparable coordination patterns.
- **Evaluate lexical features through a downstream sentence-level task.** Intersections of feature sets are compared directly against each other. This shows which lexical properties actually help, rather than relying only on intrinsic lexicon accuracy.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Columbia University](https://md.research.elamur.ai/free/en/universities/columbia-university/) — University, New York, United States · 149 works · 48,852 citations
- [New Mexico State University](https://md.research.elamur.ai/free/en/universities/new-mexico-state-university/) — University, Las Cruces, United States · 6 works · 5,598 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works

## Cite

Cite as: Vasileios Hatzivassiloglou and Janyce M. Wiebe. 2000. Effects of Adjective Orientation and Gradability on Sentence Subjectivity. COLING 2000 Volume 1: The 18th International Conference on Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{hatzivassiloglou2000effects,
  title = {{Effects of Adjective Orientation and Gradability on Sentence Subjectivity}},
  author = {Vasileios Hatzivassiloglou and Janyce M. Wiebe},
  booktitle = {COLING 2000 Volume 1: The 18th International Conference on Computational Linguistics},
  year = {2000},
  doi = {10.3115/990820.990864},
  url = {https://doi.org/10.3115/990820.990864}
}
```

- [← previous Discriminative Reranking for Natural Language Parsing](https://md.research.elamur.ai/free/en/papers/collins-2000-discriminative-reranking-natural-language-parsing/)
- [next → Advances in domain independent linear text segmentation](https://md.research.elamur.ai/free/en/papers/choi-2000-advances-domain-independent-linear-text/)
- [↑ back to papers of 2000](https://md.research.elamur.ai/free/en/papers/2000/)

## References

19 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Douglas M. Bates and Donald G. Watts. 1988. Nonlinear Regression Analysis and its Applications. Wiley, New York. [find](https://scholar.google.com/scholar?q=Douglas+M%2E+Bates+and+Donald+G%2E+Watts%2E+1988%2E+Nonlinear+Regression+Analysis+and+its+Applications%2E+Wiley%2C+New+York%2E)
2. Edwin L. Battistella. 1990. Markedness: The Evaluative Superstructure of Language. State University of New York Press, Albany, New York. [find](https://scholar.google.com/scholar?q=Edwin+L%2E+Battistella%2E+1990%2E+Markedness%3A+The+Evaluative+Superstructure+of+Language%2E+State+University+of+New+York+Press%2C+Albany%2C+New+York%2E)
3. Rebecca Bruce and Janyce Wiebe. 2000. Recognizing subjectivity: A case study of manual tagging. Natural Language Engineering, 6(2). [find](https://scholar.google.com/scholar?q=Rebecca+Bruce+and+Janyce+Wiebe%2E+2000%2E+Recognizing+subjectivity%3A+A+case+study+of+manual+tagging%2E+Natural+Language+Engineering%2C+6%282%29%2E)
4. Kenneth W. Church. 1988. A stochastic parts program and noun phrase parser for unrestricted text. In Proceedings of the Second Conference on Applied Natural Language Processing (ANLP-88), pages 136–143, Austin, Texas, February. Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/church-1988-stochastic-parts-program-noun-phrase/)
5. Joseph L. Fleiss. 1981. Statistical Methods for Rates and Proportions. Wiley, New York, 2nd edition. [find](https://scholar.google.com/scholar?q=Joseph+L%2E+Fleiss%2E+1981%2E+Statistical+Methods+for+Rates+and+Proportions%2E+Wiley%2C+New+York%2C+2nd+edition%2E)
6. Vasileios Hatzivassiloglou and Kathleen R. McKeown. 1997. Predicting the semantic orientation of adjectives. In Proceedings of the 35th Annual Meeting of the ACL and the 8th Conference of the European Chapter of the ACL, pages 174–181, Madrid, Spain, July. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Vasileios+Hatzivassiloglou+and+Kathleen+R%2E+McKeown%2E+1997%2E+Predicting+the+semantic+orientation+of+adjectives%2E+In+Proceedings+of+the+35th+Annual+Meeting+of+the+ACL+and+the+8th+Conference+of+the+European)
7. Eduard Hovy. 1998. Combining and standardizing large-scale practical ontologies for machine translation and other uses. In Proceedings of the 1st International Conference on Language Resources and Evaluation (LREC), Granada, Spain. [find](https://scholar.google.com/scholar?q=Eduard+Hovy%2E+1998%2E+Combining+and+standardizing+large%2Dscale+practical+ontologies+for+machine+translation+and+other+uses%2E+In+Proceedings+of+the+1st+International+Conference+on+Language+Resources+and+Eva)
8. Jerrold J. Katz. 1972. Semantic Theory. Harper and Row, New York. [find](https://scholar.google.com/scholar?q=Jerrold+J%2E+Katz%2E+1972%2E+Semantic+Theory%2E+Harper+and+Row%2C+New+York%2E)
9. Kevin Knight and Steve K. Luk. 1994. Building a large-scale knowledge base for machine translation. In Proceedings of the 12th National Conference on Artificial Intelligence (AAAI-94), volume 1, pages 773–778, Seattle, Washington, July–August. American Association for Artificial Intelligence. [find](https://scholar.google.com/scholar?q=Kevin+Knight+and+Steve+K%2E+Luk%2E+1994%2E+Building+a+large%2Dscale+knowledge+base+for+machine+translation%2E+In+Proceedings+of+the+12th+National+Conference+on+Artificial+Intelligence+%28AAAI%2D94%29%2C+volume+1%2C+pages)
10. Adrienne Lehrer. 1974. Semantic Fields and Lexical Structure. North Holland, Amsterdam and New York. [find](https://scholar.google.com/scholar?q=Adrienne+Lehrer%2E+1974%2E+Semantic+Fields+and+Lexical+Structure%2E+North+Holland%2C+Amsterdam+and+New+York%2E)
11. John Lyons. 1977. Semantics, volume 1. Cambridge University Press, Cambridge, England. [find](https://scholar.google.com/scholar?q=John+Lyons%2E+1977%2E+Semantics%2C+volume+1%2E+Cambridge+University+Press%2C+Cambridge%2C+England%2E)
12. K. Mahesh and S. Nirenburg. 1995. A situated ontology for practical NLP. In Proceedings of the Workshop on Basic Ontological Issues in Knowledge Sharing, 14th International Joint Conference on Artificial Intelligence (IJCAI-95), Montréal, Canada, August. [find](https://scholar.google.com/scholar?q=K%2E+Mahesh+and+S%2E+Nirenburg%2E+1995%2E+A+situated+ontology+for+practical+NLP%2E+In+Proceedings+of+the+Workshop+on+Basic+Ontological+Issues+in+Knowledge+Sharing%2C+14th+International+Joint+Conference+on+Artific)
13. Mitchell P. Marcus, Beatrice Santorini, and Mary Ann Marcinkiewicz. 1993. Building a large annotated corpus of English: the Penn Treebank. Computational Linguistics, 19(2):313–330, June. [our card](https://md.research.elamur.ai/free/en/papers/marcus-1993-building-large-annotated-corpus-english/)
14. Randolph Quirk, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. 1985. A Comprehensive Grammar of the English Language. Longman, London and New York. [find](https://scholar.google.com/scholar?q=Randolph+Quirk%2C+Sidney+Greenbaum%2C+Geoffrey+Leech%2C+and+Jan+Svartvik%2E+1985%2E+A+Comprehensive+Grammar+of+the+English+Language%2E+Longman%2C+London+and+New+York%2E)
15. Thomas J. Santner and Diane E. Duffy. 1989. The Statistical Analysis of Discrete Data. Springer-Verlag, New York. [find](https://scholar.google.com/scholar?q=Thomas+J%2E+Santner+and+Diane+E%2E+Duffy%2E+1989%2E+The+Statistical+Analysis+of+Discrete+Data%2E+Springer%2DVerlag%2C+New+York%2E)
16. Edward Sapir. 1944. On grading: A study in semantics. Philosophy of Science, 2:93–116. Reprinted in (Sapir, 1949). [find](https://scholar.google.com/scholar?q=Edward+Sapir%2E+1944%2E+On+grading%3A+A+study+in+semantics%2E+Philosophy+of+Science%2C+2%3A93%E2%80%93116%2E+Reprinted+in+%28Sapir%2C+1949%29%2E)
17. Edward Sapir. 1949. Selected Writings in Language, Culture and Personality. University of California Press, Berkeley, California. Edited by David G. Mandelbaum. [find](https://scholar.google.com/scholar?q=Edward+Sapir%2E+1949%2E+Selected+Writings+in+Language%2C+Culture+and+Personality%2E+University+of+California+Press%2C+Berkeley%2C+California%2E+Edited+by+David+G%2E+Mandelbaum%2E)
18. John M. Sinclair (editor in chief). 1987. Collins COBUILD English Language Dictionary. Collins, London. [find](https://scholar.google.com/scholar?q=John+M%2E+Sinclair+%28editor+in+chief%29%2E+1987%2E+Collins+COBUILD+English+Language+Dictionary%2E+Collins%2C+London%2E)
19. J. Wiebe, R. Bruce, and T. O’Hara. 1999. Development and use of a gold standard data set for subjectivity classifications. In Proceedings of the 37th Annual Meeting of the Association for Computational Linguistics (ACL-99), pages 246–253, University of Maryland, June. [our card](https://md.research.elamur.ai/free/en/papers/wiebe-1999-development-use-gold-standard-data/)

*Updated 24 October 2026*