The Framework of Multiword Expression in Indonesian Language

18670

Authors

nan

Abstract

This paper presents our attempt to develop an Indonesian multi-word expression (MWE) identification framework. The framework con- sists of three different steps. In the first step, we surveyed any definitions and categorizations of MWEs in the previous studies. In the second step, sentences from our language corpora are segmented and high-frequency n-grams are ex- tracted using statistical methods. The extrac- tion results which consist of word sequences are evaluated and reorganized by using the phraseological analysis procedure. This proce- dure consists of polylexicality, fixedness, and idiomaticity. The final step is to recategorize and reevaluate data collected in the second step. In this research, data is collected from newspa- pers, Wikipedia, and scientific papers. The re- sult shows that in terms of Indonesian MWEs, despite polylexicality and fixedness, idiomatic- ity should be correlated with compositionality to get a better classification of MWEs. 1

Keywords

nan

Full Content

The Framework of Multiword Expression in Indonesian Language Totok Suhardijanto Department of Linguistics Faculty of Humanities Universitas Indonesia totok.suhardijanto@ui.ac.id Zahroh Nuriah Department of Linguistics Faculty of Humanities Universitas Indonesia zahroh.nuriah@ui.ac.id Rahmad Mahendra Faculty of Computer Science Universitas Indonesia Depok, Indonesia rahmad.mahendra@cs.ui.ac.id Adi Budiwiyanto Department of Linguistics Faculty of Humanities Universitas Indonesia adi.budiwiyanto71@ui.ac.id Abstract This paper presents our attempt to develop an Indonesian multi-word expression (MWE) identification framework. The framework con- sists of three different steps. In the first step, we surveyed any definitions and categorizations of MWEs in the previous studies. In the second step, sentences from our language corpora are segmented and high-frequency n-grams are ex- tracted using statistical methods. The extrac- tion results which consist of word sequences are evaluated and reorganized by using the phraseological analysis procedure. This proce- dure consists of polylexicality, fixedness, and idiomaticity. The final step is to recategorize and reevaluate data collected in the second step. In this research, data is collected from newspa- pers, Wikipedia, and scientific papers. The re- sult shows that in terms of Indonesian MWEs, despite polylexicality and fixedness, idiomatic- ity should be correlated with compositionality to get a better classification of MWEs. 1 Introduction Since Firth (1957) stated that in language commu- nication, meaning is usually conveyed by word group, many scholars had explored this idea and had ended up with a concept of regularly significant fre- quent sequences of words called multiword expres- sions. Different scholars used different terms for these sequential linguistic phenomena (see Burger, 2015; Fleischer, 1997; Sprenger, 2003; Biber & Conrad, 2019). Burger (2015) and Fleischer (1997) mentioned it as phraseology, while Sprenger (2003) and Bider & Conrad (2019), called these word se- quences as fixed expressions. Burger (2015) distin- guishes phraseological units or MWEs by three fea- tures: polylexicality, fixedness, and idiomaticity, where idiomaticity need not be present in all phra- seological units. Up to now, in the field of linguistics and lan- guage studies, the classification of MWEs is still considered as a challenging task. Even more so, as there is no general consensus about what counts as an MWE. Masini (2005: 145) viewed MWEs as “lexical units larger than a word that can bear both idiomatic and compositional meanings. (…) the term multi-word expression is used as a pre-theoret- ical label to include the range of phenomena that goes from collocations to fixed expressions”. In a more detailed way, Sailer & Markantonatou (2018: v) defines MWEs as ‘any expression that contains more than one basic lexical element and that is lex- icalised, fixed, idiomatic, or irregular in one way or the other.’ For both the natural language application and the linguistic theory, Multiword expressions (MWEs) are challenging because they are often difficult to be classified by an application of the machinery devel- oped for free combinations where the default is that the meaning of an utterance can be predicted from its structure. Compared to MWEs in European lan- guages, there are only a few number of primarily descriptive works on MWEs …

Content truncated to 500 words

Labeled Text

This paper presents our attempt to develop an Indonesian multi-word expression (MWE) identification framework. The framework consists of three different steps. In the first step, we surveyed any definitions and categorizations of MWEs in the previous studies. In the second step, sentences from our language corpora are segmented and high-frequency n-grams are extracted using statistical methods. The extraction results which consist of word sequences are evaluated and reorganized by using the phraseological analysis procedure. This procedure consists of polylexicality, fixedness, and idiomaticity. The final step is to recategorize and reevaluate data collected in the second step. In this research, data is collected from newspapers, Wikipedia, and scientific papers. The result shows that in terms of Indonesian MWEs, despite polylexicality and fixedness, idiomaticity should be correlated with compositionality to get a better classification of MWEs. 1

Publication Details

Paper ID:
18670
Year:
2020
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan