Panuring pampanitikan (PanPam): A sentence syntax and semantic based grammar checker for Filipino

In today content, people use the computer to type articles Filipino articles. Instead of manually checking for errors, a system can check it. One application of Natural Language Processing (NLP) is grammar checking. The general notion is that grammar checkers, by virtue of their name, attempt to loc...

Full description

Saved in:
Bibliographic Details
Main Authors: Jasa, Michael A., Palisoc, Justin O., Villa, Martee M.
Format: text
Language:English
Published: Animo Repository 2007
Subjects:
Online Access:https://animorepository.dlsu.edu.ph/etd_bachelors/11049
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: De La Salle University
Language: English
Description
Summary:In today content, people use the computer to type articles Filipino articles. Instead of manually checking for errors, a system can check it. One application of Natural Language Processing (NLP) is grammar checking. The general notion is that grammar checkers, by virtue of their name, attempt to locate errors, mark the portion of the text as dubious or erroneous, and give feedback to the users. In connection with this, this research developed a Filipino sentence syntax and semantic checker that was integrated in an open-source word processing tool. Panuring Pampanitikan is a plug-in to OpenOffice.org. Write that enables the said word processor to check for Filipino morphology, syntax and semantics. Testing of syntax checking yielded a recall of 90.47% and precision of 73.07% using 81 sentences. Testing of the spell checking using 350 misspelled words showed that the spell checker can suggest the expected word from misspelled words with one edit distance 99.33% of the time, and 42% of the time for words with two edit distances 350 misspelled words were used in spell check testing. 99% of the 150% misspelled words with one edit distance had the expected word suggestion, while 42% of the 200 misspelled words with two edit distances had the expected word suggestion. Semantic checking has an accuracy score of 80%.