Indonesian Sentence Parser

Client: Group Project

Role: Algorithm & Logic Developer

Tech Stack: Python

Indonesian Sentence Parser

Project Overview

A syntactic parser that validates whether an Indonesian sentence follows standard active news structures: S-P, S-P-O, S-P-K, or S-P-O-K (Subject–Predicate–Object–Adverb). The system uses a predictive parsing table derived from a context-free grammar to accept or reject input sentences.

Problem Statement

Natural language processing for Indonesian is underdeveloped compared to English. This project addresses the need for educational tools that help learners understand basic sentence construction in Indonesian by providing instant grammatical validation—a useful aid for language learners and NLP beginners.

How It Is Done

We defined a formal grammar with terminals (e.g., 'Ana', 'membeli', 'di pasar') and non-terminals (S, P, O, K). A parsing table was constructed to guide top-down predictive parsing. The program reads input tokens and uses stack-based parsing to validate structure. Error handling provides clear feedback for invalid constructions. Though simple, it demonstrates core compiler design and linguistic analysis principles.