Initial Experiments in Data-Driven Morphological Analysis for Finnish

Proceedings of the 4th International Workshop for Computational Linguistics for Uralic Languages (IWCLUL 2018), pages 100–107, Helsinki, Finland, January 8–9, 2018. c⃝2018 Association for Computational Linguistics 100 Initial Experiments in Data-Driven Morphological Analysis for Finnish Miikka Silfverberg University of Colorado Boulder Department of Linguistics miikka.silfverberg@colorado.edu Mans Hulden University of Colorado Boulder Department of Linguistics mans.hulden@colorado.edu Abstract This paper presents initial experiments in data-driven morphological analy- sis for Finnish using deep learning methods. Our system uses a character based bidirectional LSTM and pretrained word embeddings to predict a set of morpho- logical analyses for an input word form. We present experiments on morpholog- ical analysis for Finnish. We learn to mimic the output of the OMorFi analyzer on the Finnish portion of the Universal Dependency treebank collection. The re- sults of the experiments are encouraging and show that the current approach has potential to serve as an extension to existing rule-based analyzers. Tiivistelmä Esittelemme kokeita aineistolähtöisellä syväoppimismenetelmiin perustuval- la suomen kielen morfologisella analysaattorilla. Esittelemämme järjestelmä pe- rustuu merkkipohjaisiin LSTM-malleihin ja esiopetettuihin sanaupotuksiin. Jär- jestelmämme oppii matkimaan OMorFi-jäsennintä, joka on suomen kielen morfo- loginen analysaattori. Teemme kokeita Universal Dependency -puupankkikoko- elman suomenkielisellä osuudella. Kokeemme osoittavat, että koneoppimismene- telmät tarjoavat lupaavan lähestymistavan suomen kielen morfologiseen analyy- siin. 1 Introduction The task of morphological analysis consists of providing a word form with the com- plete set of morphological readings it can attain (see Figure 1). It is a cornerstone in the development of natural language processing (NLP) utilities for morphologically complex languages such as the Uralic languages. It is a necessary preprocessing task because of the high type-to-token ratio, which is prevalent in morphologically com- plex languages. Additionally, phenomena like compounding and derivation, which frequently produce previously unseen lexemes, necessitate the use of morphological analyzers. 101 tunne tunne Noun+Sg+Nom tuntea Verb+Act+Impv+Sg2 tuntea Verb+Act+Indv+Pres+Con Figure 1: A complete set of morphological readings for the Finnish word tunne. Hand-crafted analyzers (Koskenniemi, 1983) are the gold standard for morpho- logical analysis. Creation of such analyzers is, however, a labor intensive process and requires expertise in linguistics, the target language and the rule formalisms used to create these analyzers. Moreover, analyzers need to be continuously updated with new lexemes in order to maintain high coverage on running text. In this paper, we investigate an alternative to hand-crafted analyzers, namely, data-driven morphological analyzers which are learned from annotated training data. In our case, the training data consists of words and complete sets of analyses. During test time, the system takes a Finnish word such as kisaan (‘into the competition’ or ‘I am competing’) as input and gives a set of analyses {Noun+Sg+Ill, Verb+Act+Indv+Pres+Sg1} as output. We present experiments in data-driven morphological analysis of Finnish. We learn to mimic the OMorFi analyzer (Pirinen et al., 2017) on the Finnish portion of the Universal Dependency treebank collection (Pyysalo et al., 2015). The data sets and OMorFi analyzer are further discussed in Section 3. We use a deep learning model encompassing a character-level recurrent model, which maps words onto sets of anal- …

Content truncated to 500 words