English Phrasal Verbs: from Lexicon-Grammar to Natural Language Processing

Résumé En

As Sag et al. (2002: 14) state, multiword expressions "constitute a key problem that must be resolved in order for linguistically precise NLP [Natural Language Processing] to succeed." This article presents the results of using the linguistic development environment, NooJ, together with manually constructed lexicon-grammar tables of transitive and neutral English phrasal verbs, to automatically recognize these constructions with and without insertion in large corpora. Our analysis confirms that NooJ is a very powerful tool for parsing compound expressions that involve insertion, such as English phrasal verbs. Although the overall accuracy was 85%, the concordances generated from the most exhaustive tables (the particles up and out) achieved 97% accuracy, which confirms that listing all phrasal verbs in a lexicon-grammar database can help distinguish prepositional usage from true particles in NLP. Accuracy, as well as recall, can be improved in the future by adding intransitive phrasal verbs, phrasal prepositional verbs, and other multiword idioms to the NooJ database.