An Expert Lexicon Approach to Identifying English Phrasal Verbs

What it is about

Phrasal verbs (look for, turn on, blow up) sit between lexicon and syntax, and a plain lexical lookup mistags them both ways: it misses separable uses such as "turn the radio off", and it fires on "looked for quite a while", where "for" starts a prepositional phrase. The authors put a phrasal-verb "expert lexicon" between a shallow parser (verb groups, base NPs, named entities) and a deep parser. Entries are written with macros for the three verb-particle types, and each macro carries constraints such as "no NP after the particle" and "not passive". On a 357-line informal corpus of 347 phrasal verbs, the method reaches 95.8% F-score, against 86.9% for the bare lookup and 91.6% for a baseline that skips inserted pronouns and adverbs. On three verbs (look for, turn on, blow up) over 23,557 news articles it reaches 95.8% to 97.5%. An indexed lexicon runs about 30 times faster than a sequential finite-state pass.

Ideas

• I1. 13. Verbs and 19. The lexicon: vocabulary, multiword expressions and register: write the particle verbs as macros, one per type (V+P, V+NP+P, V+P+NP) with the constraint "no NP after the particle" as a rule, not as a list. The macro gives the lexicon one line per verb, and the rule is explainable.
• I2. en/src/morph.rs and the tagger output: after the tagger, mark a particle as bound to its verb only when the post-particle constraint holds. Keep a reason string per decision ("turn ... off: NP inserted"), so every phrasal-verb decision can say why.
• I3. A negative control: build a set of sentences with the same verb-particle strings in both readings (looked for you / looked for a while) and require the rule to score worse on the shuffled or the wrong-particle version.
• I4. Measure with the paper's split: correct, missing and spurious counts, precision and recall, on a held-out informal set written by us, and report the bare-lookup baseline next to it.

For Mova

The lexicon work belongs in 19. The lexicon: vocabulary, multiword expressions and register and 13. Verbs. The code points are the lexicon and the dictionary in en/src/dict.rs, the tokenizer en/src/tok.rs, and the parse-side check in en/src/parse.rs. The measure is precision and recall of phrasal-verb identification on a held-out set kept out of the lexicon development, run twice for determinism, with the bare lookup as the baseline and a shuffled-particle negative control.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.3115/1075096.1075161

@inproceedings{li2003expert,
  title = {{An Expert Lexicon Approach to Identifying English Phrasal Verbs}},
  author = {Li, Wei and Zhang, Xiuhong and Niu, Cheng and Jiang, Yuankai and Srihari, Rohini K.},
  booktitle = {Proceedings of the 41st Annual Meeting of the Association for Computational Linguistics},
  year = {2003},
  doi = {10.3115/1075096.1075161},
  url = {https://doi.org/10.3115/1075096.1075161}
}

References

13 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Breidt. E., F. Segond and G. Valetto. 1994. Local Grammars for the Description of Multi-Word Lexemes and Their Automatic Recognition in Text. Proceedings of Comlex-2380 - Papers in Computational Lexicography, Linguistics Institute, HAS, Budapest, 19-28. find
2. Breidt, et al. 1996. Formal description of Multi-word Lexemes with the Finite State formalism: IDAREX. Proceedings of COLING 1996, Copenhagen. find
3. Bolinger, D. 1971. The Phrasal Verb in English. Cambridge, Mass., Harvard University Press. find
4. Church, K. 1988. A stochastic parts program and noun phrase parser for unrestricted text. Proceedings of ANLP 1988. find
5. Di Sciullo, A.M. and E. Williams. 1987. On The Definition of Word. The MIT Press, Cambridge, Massachusetts. find
6. Fraser, B. 1976. The Verb Particle Combination in English. New York: Academic Press. find
7. Pelli, M. G. 1976. Verb Particle Constructions in American English. Zurich: Francke Verlag Bern. find
8. Sag, I., T. Baldwin, F. Bond, A. Copestake and D. Flickinger. 2002. Multiword Expressions: A Pain in the Neck for NLP. Proceedings of CICLING 2002, Mexico City, Mexico, 1-15. find our card
9. Shaked, N. 1994. The Treatment of Phrasal Verbs in a Natural Language Processing System, Dissertation, CUNY. find
10. Silberztein, M. 2000. INTEX: An FST Toolbox. Theoretical Computer Science, Volume 231(1): 33-46. find
11. Small, S. and C. Rieger. 1982. Parsing and comprehending with word experts (a theory and its realisation). W. Lehnert and M. Ringle, editors, Strategies for Natural Language Processing. Lawrence Erlbaum Associates, Hillsdale, NJ. find
12. Srihari, R., W. Li, C. Niu and T. Cornell. 2003. InfoXtract: An Information Discovery Engine Supported by New Levels of Information Extraction. Proceeding of HLT-NAACL Workshop on Software Engineering and Architecture of Language Technology Systems, Edmonton, Canada. find
13. Villavicencio, A. and A. Copestake. 2002. Verb-particle constructions in a computational grammar of English. Proceedings of the Ninth International Conference on Head-Driven Phrase Structure Grammar, Seoul, South Korea. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/li-2003-expert-lexicon-approach-identifying-english/