SIL International Home

The Linguist's Shoebox

Integrated data management and analysis for the field linguist


You can write word formulas to eliminate invalid parses.

When Shoebox produces both valid and invalid parses for a word because of limitations in its basic morphological parsing features, it can use "word formulas" to select the valid parses automatically. A common problem in Shoebox parsing is invalid sequences of morphemes (especially involving wrong homonyms). You can solve it by writing "formulas" that define valid sequences of morphemes (usually in terms of part of speech categories or glosses in lexical databases). Word formulas can eliminate invalid parses, but they don't enable any additional valid parses. Note that Shoebox doesn't have "sentence formulas" that would use a word's surrounding context to resolve ambiguities in its meaning when there are multiple valid parses. For more information: In Shoebox Help, search for Word Formulas.

Index of tips: homonyms; parsing; word formulas
List of tips