A Finite-State Morphological Analyser for Sindhi

A Finite-State Morphological Analyser for Sindhi Raveesh Motlani1, Francis M. Tyers2 and Dipti M. Sharma1 1FC Kohli Center on Intelligent Systems (KCIS), International Institute of Information Technology Hyderabad, Telangana, India 2 HSL-fakultehta, UiT Norgga árktalaš universitehta N-9019 Romsa raveesh.motlani@research.iiit.ac.in, francis.tyers@uit.no, dipti@iiit.ac.in Abstract Morphological analysis is a fundamental task in natural-language processing, which is used in other NLP applications such as part-of-speech tagging, syntactic parsing, information retrieval, machine translation, etc. In this paper, we present our work on the development of free/open-source finite-state morphological analyser for Sindhi. We have used Apertium’s lttoolbox as our finite-state toolkit to implement the transducer. The system is developed using a paradigm-based approach, wherein a paradigm defines all the word forms and their morphological features for a given stem (lemma). We have evaluated our system on the Sindhi Wikipedia, which is a freely-available large corpus of Sindhi and achieved a reasonable coverage of about 81% and a precision of over 97%. Keywords: Sindhi, Morphological Analysis, Finite-State Machines 1. Introduction Morphology describes the internal structure of words in a language. A morphological analysis of a word involves describing one or more of its properties such as: gender, number, person, case, lexical category, etc. Morphological analysis of a word thus becomes a fundamental and cru- cial task in natural-language processing for any language. Morphological analysis is vital in various NLP applications such as spell-checking, part-of-speech tagging, dependency parsing, information retrieval, machine translation, etc. Sindhi is an Indo-Aryan language with an inflectional mor- phology. The language can be written in one of two writ- ing systems, either in the Devanagari writing system or the Perso-Arabic writing system. In this paper we will dis- cuss our approach in building a morphological analyser for Sindhi in Perso-Arabic script. The paper is organised in the following manner. We give an overview of the language in Section 2. We have discussed the work done previously in Section 3. We have described our developmental process in Section 4. The evaluation of our system is described in Section 5. The future work is dis- cussed in Section 6. The concluding remarks are given in Section 7. 2. The linguistic properties of Sindhi Sindhi belongs to the Indo-Aryan language family. It has a long history and its origins can be traced back to 1500 BCE. The name Sindhi is derived from Sindhu which was the local name for the Indus river. The Sindh region has gone through various invasions and as a result the vocabulary contains many loan words from Persian, Arabic, Hindi and Urdu. Sindhi is an official language in both Pakistan and India. Estimated number of native speakers are approximately 59 million in these two countries (Khubchandani, 2003). How- ever, it is also spoken by people in various other countries. Some of the linguistic properties of Sindhi are discussed be- low. ڱ[ŋ]ڃ[ɲ]ٻ[ɓ] ڳ[ɠ]ڄ[ʄ]ڀ[bʱ] ڪ[k]ڇ[cʰ]ڌ[dʱ] ڻ[ɳ]ٺ[ʈʰ]ڏ[ɗ] ڦ[pʰ]ٽ[ʈ]ڊ[ɖ] ڙ[ɽ]ٿ[tʰ]ڍ[ɖʱ] Table 1: The characters in the Sindhi alphabet which are not found in the Persian alphabet and their phonetic value. 2.1. Orthography Historically, Sindhi has been written …

Content truncated to 500 words