Automatic Extraction of Typological Linguistic Features from Descriptive Grammars

Abstract

Linguistic typology is an area of linguistics concerned with analysis of and comparison between natural languages of the world based on their certain linguistic features. For that purpose, historically, the area has relied on manual extraction of linguistic feature values from textural descriptions of languages. This makes it a laborious and time expensive task and is also bound by human brain capacity. In this study, we present a deep learning system for the task of automatic extraction of linguistic features from textual descriptions of natural languages. First, textual descriptions are manually annotated with special structures called semantic frames. Those annotations are learned by a recurrent neural network, which is then used to annotate un-annotated text. Finally, the annotations are converted to linguistic feature values using a separate rule based module. Word embeddings, learned from general purpose text, are used as a major source of knowledge by the recurrent neural network. We compare the proposed deep learning system to a previously reported machine learning based system for the same task, and the deep learning system wins in terms of F1 scores with a fair margin. Such a system is expected to be a useful contribution for the automatic curation of typological databases, which otherwise are manually developed.

Coding culture: challenges and recommendations for comparative cultural databases

“…However, they are often extremely time-consuming to design, and require the data they analyse to be cleaned and formatted in particular ways, which can also be quite labour-intensive. Moreover, currently available techniques have proven to be ineffective even for tasks that might seem most amenable to such an approach, such as the extraction of typological linguistic information from published grammars (Virk et al, 2017 ). This may change in coming decades as artificial intelligence becomes more powerful and flexible (Virk et al, 2017 ), but for the moment, the need for human coders is unavoidable even when it comes to these more tractable cases.…”

A Deep Learning System for Automatic Extraction of Typological Linguistic Information from Descriptive Grammars

“…Previously, a few experimental techniques and associated systems have been reported for automatic extraction of typological information. In (Borin et al, 2018;Virk et al, 2017), the authors have reported on simple pattern matching and syntactic parsing based systems. The systems have modest accuracy and recall and are very restricted with respect to the number of features they can target.…”

Term Spotting: A Quick-and-dirty Method for Extracting Typological Features of Language from Grammatical Descriptions

“…Given its novelty, only a few embryonic approaches (Virk et al, 2019;Wichmann and Rama, 2019;Macklin-Cordes et al, 2017;Hammarström, 2013;Virk et al, 2017) have addressed the task so far. Of these, some are word-based and some combine words with more elaborate analyses of the source texts such as frame-semantics (Virk et al, 2019).…”