Developing an Android Application for Analyzing Indonesian Syntax: A Rule and Probability-Based POS Tagging Approach

Abstract

The investigation of the grammatical level of syntax, particularly Indonesian, focuses exclusively on sentence formation and does not include the corpus. This renders Indonesian grammatical corpus data less relevant in corpus-based grammatical investigations. The study offers a detailed overview of the utilization of Android applications based on POS Tagging data. The method in this study was qualitative focusing on the development of an application that utilizes Rule and Probability-based POS Tagging data from Leipzig Indonesian Mix_2013 to determine the categories, functions, and roles of Indonesian syntax with lexical categories including V (copula, existence, and equative) as the predicate potential on function. The application was designed to be compatible with the Android system by Integrating POS tagging into the System Development Life Cycle (SDLC), enabling wider accessibility to a larger user base. The result of this research introduces a program designed as a tool to search syntactic categories in Indonesian. The program uses a sequential search technique, which is a linear search method, to make it easier for users to find specific syntactic functions. By applying syntactic categories and functions using POS Tagging data from the Leipzig Indonesian Mix_2013 corpus, the study achieved significant insights into the roles of Indonesian syntax. POS Tagging based on the generated rules and probabilities achieves an accuracy rate of 92.53% for category tags and functions.

Keywords

• android applications
• digital language learning
• grammatical corpus
• indonesian syntax
• lexical categories
• pos tagging