Using Machine Learning to Maintain Rule-based Named-Entity Recognition and Classification Systems

Using Machine Learning to Maintain Rule-based Named-Entity Recognition and Classification Systems Georgios Petasis †, Frantz Vichot §, Francis Wolinski § Georgios Paliouras †, Vangelis Karkaletsis † and Constantine D. Spyropoulos † † Institute of Informatics and Telecommunications, National Centre for Scientific Research “Demokritos”, 153 10 Ag. Paraskevi, Athens, Greece § Informatique-CDC 4, rue Berthollet 94114 Arcueil, France {petasis,paliourg,vangelis,costass}@iit.demokritos.gr {frantz.vichot, francis.wolinski}@caissedesdepots.fr Abstract This paper presents a method that as- sists in maintaining a rule-based named-entity recognition and classifi- cation system. The underlying idea is to use a separate system, constructed with the use of machine learning, to monitor the performance of the rule-based sys- tem. The training data for the second system is generated with the use of the rule-based system, thus avoiding the need for manual tagging. The dis- agreement of the two systems acts as a signal for updating the rule-based sys- tem. The generality of the approach is illustrated by applying it to large cor- pora in two different languages: Greek and French. The results are very en- couraging, showing that this alternative use of machine learning can assist sig- nificantly in the maintenance of rule- based systems. 1 Introduction Machine learning has recently been proposed as a promising solution to a major problem in lan- guage engineering: the construction of lexical resources. Most of the real-world language en- gineering systems make use of a variety of lexi- cal resources, in particular grammars and lexi- cons. The use of general-purpose resources is ineffective, since in most applications a special- ised vocabulary is used, which is not supported by general-purpose lexicons and grammars. For this reason, significant effort is currently put into the construction of generic tools that can quickly adapt to a particular thematic domain. The adaptation of these tools mainly involves the adaptation of domain-specific semantic lexi- cal resources. Named-entity recognition and classification (NERC) is the identification of proper names in text and their classification as different types of named entity (NE), e.g. persons, organisations, locations, etc. This is an important subtask in most language engineering applications, in par- ticular information retrieval and extraction. The lexical resources that are typically included in a NERC system are a lexicon, in the form of gaz- etteer lists, and a grammar, responsible for rec- ognising the entities that are either not in the lexicon or appear in more than one gazetteer lists. The manual adaptation of those two re- sources to a particular domain is time- consuming and in some cases impossible, due to the lack of experts. The exploitation of learning techniques to support this adaptation task has attracted the attention of researchers in language engineering. However, the adaptation of lexical resources to a specific domain at a certain point in time is not sufficient on its own. The performance of a NERC system degrades over time (Vichot et al., 1999; Wolinski et al., 2000) due to the introduc- tion of new NEs or the change in the meaning of existing ones. We need to find ways that facili- …

Content truncated to 500 words