A Morphological Lexicon of Esperanto with Morpheme Frequencies

A Morphological Lexicon of Esperanto with Morpheme Frequencies Eckhard Bick University of Southern Denmark Campusvej 55, DK-5230 Odense M Email: eckhard.bick@mail.dk Abstract This paper discusses the internal structure of complex Esperanto words (CWs). Using a morphological analyzer, possible affixation and compounding is checked for over 50,000 Esperanto lexemes against a list of 17,000 root words. Morpheme boundaries in the resulting analyses were then checked manually, creating a CW dictionary of 28,000 words, representing 56.4% of the lexicon, or 19.4% of corpus tokens. The error percentage of the EspGram morphological analyzer for new corpus CWs was 4.3% for types and 6.4% for tokens, with a recall of almost 100%, and wrong/spurious boundaries being more common than missing ones. For pedagogical purposes a morpheme frequency dictionary was constructed for a 16 million word corpus, confirming the importance of agglutinative derivational morphemes in the Esperanto lexicon. Finally, as a means to reduce the morphological ambiguity of CWs, we provide POS likelihoods for Esperanto suffixes. Keywords: Morphological Analysis, Esperanto, Affixation, Compounding, Morpheme Frequencies 1. Introduction As an artificial language with a focus on regularity and facilitation of language acquisition, Esperanto was designed with a morphology that allows (almost) free, productive combination of roots, affixes and inflexion endings. Thus, the root 'san' (healthy) not only accepts its prototypical adjectival ending '-a', but also other part-of- speech endings ('san|e' - healthily, 'san|on' - cheers)1, as well as prefixes ('mal+san|a' - unhealthy [mal=opposite]), suffixes ('sanigi' - cure [ig=make]) or multiple affixes ('mal+san+ul+ej|o' - hospital [ul=person, ej=place]). In addition, compounding is common ('san+serv|oj' - health services, 'san+asekur|o' - health insurance). This morphological versatility reduces the number of to-be- learned lexemes and is - together with high linguistic transparency - generally cited as a main reason for the language's easiness as an L2 and its usefulness as a propedeutic language (e.g. Telier 2013). Given the agglutinative properties of the language, it is pedagogically interesting to know which morphemes contribute most to the Esperanto lexicon, in order to teach them first. In other words, on top of basic word lists and word frequencies (Quasthoff et al. 2014), Esperanto teachers and text book authors also need morpheme frequencies. The work described here is intended to provide just that - a frequency dictionary of Esperanto morphemes, as well as the tools required to build it from corpus data. 1 The POS markers, attached agglutinatively at the end, after possible suffixes, are the following: -o = noun, -a = adjective, -e = adverb, -i = verb (infinitive). After these, a plural '-j' (for -a/o) and accusative '-n' (for -a/o/e) can be added as an inflection ending. For verbs, the infinitve ending can be replaced by a tense vowel (a = present, i = past, o = future) plus a finity ending '-s' (e.g. aĉetis [bought]). Participles are built by adding '-t' (passive) or '-nt' (active) instead (e.g. aĉetinta [having bought]). 2. Morphological analysis From a language technology perspective, inflexional regularity, morphological transparency and surface-based access to semantic features turn POS tagging of Esperanto …

Content truncated to 500 words