Filipino Language Grammar Checker Focusing on the Language Sentence Order Form Using Rule-Based POS Algorithm

This study is focused on designing a rule-based algorithm model using the Sentence Order Form of the Filipino Language for checking its grammar. The method presents a set of rules that are used to identify the Sentence Order Form of the Filipino sentences. Various sentences from different sources have been obtained and tested by the model. The algorithm performs by tagging the POS Tags, converting it into a simpler POS Tags, tagging its Subject and Predicate through Sentence Analyzer and tagging Di-Karaniwan and Karaniwan sentences through the Sentence Order Form. Through its testing, the model has performed good results, having a precision of 65 percent and a recall of 69 percent. Taglish sentences were also detected, and deep Filipino words were usually the cause of an unknown tagging, therefore resulting in a wrong grammar result. This shows that it may need improvements in checking the POS Tags, and by having more rules added to the algorithm in order to detect more different and unique sentences. Nevertheless, the study is a great addition to the improvement of NLP focusing on Filipino language.