FAQ

Common questions about Esperanto and about turning it into data for the library. Each answer links to the page where the topic is covered in full.

What is Esperanto, in one sentence?

A constructed international auxiliary language, published by L. L. Zamenhof in 1887, designed to be regular and easy to learn — with no grammatical exceptions. See History.

Why build an NLP library for Esperanto specifically?

Because its grammar is fully regular, so a rule-based tokenizer, lemmatizer and POS tagger are actually feasible — endings and affixes follow rules that never break, unlike in natural languages. See History § Why it matters here.

How do I know a word's part of speech?

From its final vowel: -o is a noun, -a an adjective, -e an adverb, -i a verb infinitive. This one rule covers most content words. See Grammar § Word-class endings.

How does lemmatization work in Esperanto?

Peel the grammatical endings and affixes off a word to reach its root, then re-attach the class vowel. malsanulejojn → strip -n, -j, -o, then -ej-, -ul-, mal- → root san-. Each step is a table lookup. See Grammar § Affixes.

What are the correlatives?

A regular 5 × 9 grid of 45 common words (this/that/which/some/every/no × thing/person/place/time…), each a prefix plus an ending. Learn the grid and you get all 45. See Grammar § Correlatives.

What is the accusative -n for?

It marks the direct object of a verb (and direction of motion). Nouns and their adjectives both take it, and it stacks after the plural -j: interesajn librojn. See Grammar § Grammatical inflections.

How should I handle the special letters ĉ ĝ ĥ ĵ ŝ ŭ in code?

Treat them as ordinary Unicode letters and normalise input to a single canonical form (NFC, real diacritics) before tokenizing, converting the x-system (cx, gx, …) or h-system if present. See Grammar § Alphabet.

What counts as a stop-word in Esperanto?

The article la, the pronouns, the correlatives, and a fixed set of prepositions, conjunctions and common adverbs — mostly closed classes, so the list is finite. This project ships 122 of them — roots only, since the inflected forms are generated by rule. See resources/README.md § Roots, not forms.

Where does the library get its word lists from?

From JSON files under resources/, which are the single source of truth: the docs render them at build time and the library will read the same files at runtime, so they can never disagree. See resources/README.md.

Where can I find real Esperanto text to test the library?

From corpora such as the Tekstaro de Esperanto, the Esperanto Wikipedia, and public-domain books on Project Gutenberg — mind each source's licence. See Resources § Corpora.

Is the grammar content here trustworthy?

It follows Ivy Kellerman Reed's A Complete Grammar of Esperanto (1910), a public-domain reference included in the repository. See Resources § Books.