Towards Computational Resource Grammars for Runyankore and Rukiga

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 2846–2854 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 2846 Towards Computational Resource Grammars for Runyankore and Rukiga David Bamutura, Peter Ljungl¨of, Peter Nabende Chalmers Univ. of Tech. & Mbarara Univ. of Sci. & Tech. , Univ. of Gothenburg , Makerere University bamutra@chalmers.se, peter.ljunglof@cse.gu.se, peter.nabende@gmail.com Abstract In this paper, we present computational resource grammars of Runyankore and Rukiga (R&R) languages. Runyankore and Rukiga are two under-resourced Bantu Languages spoken by about 6 million people indigenous to South Western Uganda, East Africa. We used Grammatical Framework (GF), a multilingual grammar formalism and a special-purpose functional programming language to formalise the descriptive grammar of these languages. To the best of our knowledge, these computational resource grammars are the first attempt to the creation of language resources for R&R. In Future Work, we plan to use these grammars to bootstrap the generation of other linguistic resources such as multilingual corpora that make use of data-driven approaches to natural language processing feasible. In the meantime, they can be used to build Computer-Assisted Language Learning (CALL) applications for these languages among others. Keywords: Grammar, Syntax, Morphology, Runyankore, Rukiga, less-resourced Languages, Grammatical Framework, Resource Grammar Library 1. Introduction Runyankore & Rukiga (hereafter R&R) are two heavily under-resourced Bantu languages. Their limited presence on the web makes it difficult to develop substantial com- putational linguistic resources for these languages. Conse- quently, the lack of such resources makes the use of data- driven Natural Language Processing (NLP) approaches un- suitable for these languages. However, rule-based ap- proaches such as grammars, can be used to bootstrap the creation of such resources. In this paper we present com- putational resource grammars of these two languages de- veloped using Grammatical Framework (GF). 1.1. Grammatical Framework (GF) GF is a multilingual grammar formalism, a logical frame- work and a special-purpose functional programming lan- guage for defining grammars of both formal and natural languages (Ranta, 2011; Ranta, 2009a). We chose GF be- cause it does not need any additional linguistic resources, and being multilingual, it can be used to develop resources for under-resourced languages by using existing linguistic resources of well-resourced languages already covered in its Resource Grammar Library (RGL) (Ranta, 2009a; Ko- lachina and Ranta, 2016). 1.2. Abstract and Concrete Syntax Each grammar in GF consists of an abstract and concrete syntax. The abstract syntax defines a set of abstract syn- tactic structures, called abstract terms or trees, which are used to define a language-independent or semantic mean- ing representation. The concrete syntax defines a relation between the abstract structures and their language-specific constructions. This makes it possible to define several sets of concrete “syntaxes” for one single abstract syntax. The single abstract syntax then acts as an interlingua between different languages. The concept of a shared abstract syn- tax is the reason for the multilingual capabilities of GF. 1.3. Resource & Application Grammars Grammars designed in GF are of two types: resource and …

Content truncated to 500 words