Version 6.3.0

Building a corpus-specific dataset

TL;DR

• you can build an initial parser from existing datasets
• modify or add vocabulary or inflectional rules by editing delimited-text files

Fuller contents

• defining an orthography (alphabet.fst)
• organizing tabular files
• tabulae’s grammatical vocabulary for each anlaytical type (“part of speech”)

---

Table of contents

• An alphabet for your parser
• Morphological rules
• Irregular forms
• The lexicon of stems
• Indeclinable stems
• Examples of stem types