This document summarizes a thesis that proposed enhancements to the NormAPI algorithm for normalizing Filipino shortcut texts. The researchers focused on improving the dictionary substitution approach in phrases and words used in the algorithm. They added processes like including word frequencies from the existing 3000-word database to prevent stagnation. They also capitalized the first letter of nouns and the first word of outputs, and included punctuation marks to improve grammar in normalized texts. Testing showed the enhanced algorithm produced more accurate normalizations of shortcut texts compared to the original NormAPI. However, there is still potential for further improvements in future enhancements of the NormAPI algorithm.
1. AN ENHANCEMENT OF THE NORMAPI ALGORITHM FOR
NORMALIZING FILIPINO SHORTCUT TEXTS
A Thesis
Presented to the
Faculty of the Computer Science Department
College of Engineering and Technology
PAMANTASAN NG LUNGSOD NG MAYNILA
(University of the City of Manila)
Intramuros, Manila
In Partial Fulfilment of the Requirements for the Degree
Bachelor of Science in Computer Studies (BSCS) Major
in Computer Science
By:
AINSLEY KAYE A. NIRONA
MARY ANGELI D. VALMONTE
2. ABSTRACT
Shortcut texting are used by almost every people around the world especially
Filipinos. And for as we all know, Filipinos are fond of communicating with each other
even for merely texting 24/7. Due to this, Philippines became known as the texting capital
of the world and is also known for using shortcut texts. It is easier and lessens the time to
type. This research involves the improvement of the “NormAPI: An API for Normalizing
Filipino Shortcut texts” (2014). NormAPI is an algorithm with different techniques
combined to transform shortcut texts in its original form. The research mainly focuses on
how the system is able to improve its performance by enhancing its algorithm. The
researchers have focused on the dictionary substitution approach in phrases and words.
The goals of the researchers were to improve the algorithm of the NormAPI and in order
to do so they have added processes in the algorithm. These processes include the addition
of the frequency in its data. The researchers included this in the algorithm due to the 3000
data in the database so that there will be movement in the data and will not be stagnant.
Then, the researchers capitalized the first letter of the nouns and first word of the output
and included punctuation marks to improve the grammar of the normalized texts.
According to the research, the researchers conclude the enhanced algorithm of NormAPI
show better results in terms of accuracy and better normalization of shortcut texts.
Although the study was a success, there is still room for improvement for future
enhancement of NormAPI Algorithm.