Building Analyses from Syntactic Inference in Local Languages: An HPSG Grammar Inference System

Kristen Howell, Emily M. Bender


Abstract
We present a grammar inference system that leverages linguistic knowledge recorded in the form of annotations in interlinear glossed text (IGT) and in a meta-grammar engineering system (the LinGO Grammar Matrix customization system) to automatically produce machine-readable HPSG grammars. Building on prior work to handle the inference of lexical classes, stems, affixes and position classes, and preliminary work on inferring case systems and word order, we introduce an integrated grammar inference system that covers a wide range of fundamental linguistic phenomena. System development was guided by 27 geneologically and geographically diverse languages, and we test the system’s cross-linguistic generalizability on an additional 5 held-out languages, using datasets provided by field linguists. Our system out-performs three baseline systems in increasing coverage while limiting ambiguity and producing richer semantic representations, while also producing richer representations than previous work in grammar inference.
Anthology ID:
2022.nejlt-1.3
Volume:
Northern European Journal of Language Technology, Volume 8
Month:
Year:
2022
Address:
Copenhagen, Denmark
Venue:
NEJLT
SIG:
Publisher:
Northern European Association of Language Technology
Note:
Pages:
Language:
URL:
https://aclanthology.org/2022.nejlt-1.3
DOI:
https://doi.org/10.3384/nejlt.2000-1533.2022.4017
Bibkey:
Cite (ACL):
Kristen Howell and Emily M. Bender. 2022. Building Analyses from Syntactic Inference in Local Languages: An HPSG Grammar Inference System. In Northern European Journal of Language Technology, Volume 8, Copenhagen, Denmark. Northern European Association of Language Technology.
Cite (Informal):
Building Analyses from Syntactic Inference in Local Languages: An HPSG Grammar Inference System (Howell & Bender, NEJLT 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2022.nejlt-1.3.pdf