Grammar Customization

Bender, Emily M.; Drellishak, Scott; Fokkens, Antske; Poulson, Laurie; Saleem, Safiyyah

doi:10.1007/s11168-010-9070-1

Cited by 33 publications

(20 citation statements)

References 47 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In particular, I used precision grammars to evaluate my results by parsing. I used the Grammar Matrix customization system (Bender et al, 2002;Bender et al, 2010) to compile precision grammars from the specifications which were output by either the baseline (Wax, 2014) or by my k-means system. In both cases, the morphotactics is represented internally as a directed acyclic graph (DAG) where nodes are affix types (position classes) and edges mean that one class serves as input to another.…”

Section: Precision Grammars and Evaluation By Parsingmentioning

confidence: 99%

“…In particular, subject and object drop are allowed in all cases, and this makes it possible to parse stand-alone verbs. Then the specification is compiled into grammar rules by the Grammar Matrix (Bender et al, 2002;Bender et al, 2010) and this grammar can be used for parsing with software such as the LKB (Copestake, 2002) or ACE (Crysmann and Packard, 2012).…”

Section: Baseline: Inferring Position Classes Dag By Input Overlapmentioning

confidence: 99%

See 1 more Smart Citation

Inferring Morphotactics from Interlinear Glossed Text: Combining Clustering and Precision Grammars

Zamaraeva

2016

Proceedings of the 14th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

In this paper I present a k-means clustering approach to inferring morphological position classes (morphotactics) from Interlinear Glossed Text (IGT), data collections available for some endangered and low-resource languages. While the experiment is not restricted to low-resource languages, they are meant to be the targeted domain. Specifically my approach is meant to be for field linguists who do not necessarily know how many position classes there are in the language they work with and what the position classes are, but have the expertise to evaluate different hypotheses. It builds on an existing approach (Wax, 2014), but replaces the core heuristic with a clustering algorithm. The results mainly illustrate two points. First, they are largely negative, which shows that the baseline algorithm (summarized in the paper) uses a very predictive feature to determine whether affixes belong to the same position class, namely edge overlap in the affix graph. At the same time, unlike the baseline method that relies entirely on a single feature, kmeans clustering can account for different features and helps discover more morphological phenomena, e.g. circumfixation. I conclude that unsupervised learning algorithms such as k-means clustering can in principle be used for morphotactics inference, though the algorithm should probably weigh certain features more than others. Most importantly, I conclude that clustering is a promising approach for diverse morphotactics and as such it can facilitate linguistic analysis of field languages.

show abstract

Section: Precision Grammars and Evaluation By Parsingmentioning

confidence: 99%

Section: Baseline: Inferring Position Classes Dag By Input Overlapmentioning

confidence: 99%

Inferring Morphotactics from Interlinear Glossed Text: Combining Clustering and Precision Grammars

Zamaraeva

2016

Proceedings of the 14th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

View full text Add to dashboard Cite

show abstract

“…The knowledge sources are the LinGO Grammar Matrix (Bender et al 2002(Bender et al , 2010, a language-independent resource which maps relatively simple linguistic descriptions to working grammar fragments, on the one hand, and IGT on the other. We aim to infer these kinds of linguistic descriptions by applying and extending the methodology of Lewis and Xia (2008), who glean information about the structure of the source languages by parsing the translation line and projecting the information through the gloss line to the source line.…”

Section: Motivation/desideratamentioning

confidence: 99%

Xigt: extensible interlinear glossed text for natural language processing

Goodman

Crowgey

Xia

et al. 2014

Lang Resources & Evaluation

Self Cite

View full text Add to dashboard Cite

This paper presents Xigt, an extensible storage format for interlinear glossed text (IGT). We review design desiderata for such a format based on our own use cases as well as general best practices, and then explore existing representations of IGT through the lens of those desiderata. We give an overview of the data model and XML serialization of Xigt, and then describe its application to the use case of representing a large, noisy, heterogeneous set of IGT.

show abstract

“…In addition, because some linguistic properties are universal (e.g., semantic compositionality or the fact that the meaning of a sentence is built up from the meaning of its parts) and others at least recur across some languages (e.g., SOV word order), it is possible to speed up the development of grammars for new languages by reusing analyses previously developed and implemented. is is the goal of the LinGO Grammar Matrix [Bender et al, 2002[Bender et al, , 2010 project, which provides a starter-kit for developing grammars compatible with DELPH-IN technology.…”

Section: Typological Databases 125mentioning

confidence: 99%

Linguistic Fundamentals for Natural Language Processing: 100 Essentials from Morphology and Syntax

Bender

2013

Synthesis Lectures on Human Language Technologies

Self Cite

View full text Add to dashboard Cite

Many NLP tasks have at their core a subtask of extracting the dependencies-who did what to whom-from natural language sentences. is task can be understood as the inverse of the problem solved in different ways by diverse human languages, namely, how to indicate the relationship between different parts of a sentence. Understanding how languages solve the problem can be extremely useful in both feature design and error analysis in the application of machine learning to NLP. Likewise, understanding cross-linguistic variation can be important for the design of MT systems and other multilingual applications. e purpose of this book is to present in a succinct and accessible fashion information about the morphological and syntactic structure of human languages that can be useful in creating more linguistically sophisticated, more languageindependent, and thus more successful NLP systems.

show abstract

Grammar Customization

Cited by 33 publications

References 47 publications

Inferring Morphotactics from Interlinear Glossed Text: Combining Clustering and Precision Grammars

Inferring Morphotactics from Interlinear Glossed Text: Combining Clustering and Precision Grammars

Xigt: extensible interlinear glossed text for natural language processing

Linguistic Fundamentals for Natural Language Processing: 100 Essentials from Morphology and Syntax

Contact Info

Product

Resources

About