Date of Publication

2009

Document Type

Master's Thesis

Degree Name

Master of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Rachel Edita O. Roxas

Defense Panel Chair

Charibeth K. Cheng

Defense Panel Member

Allan B. Borra
Rachel Edita O. Roxas

Abstract/Summary

An approach to automatic acquisition of morphological rules for Tagalog verbs is presented adopting a modified version of Moving Contracting Window Pattern Algorithm (MCWPA) (Yang, 2001) for segmenting the inflected verbs into affixes and its bases. The input to the system is a database of verbs and their corresponding inflections. 90% of this input is taken to serve as the training set and the remaining 10% to serve as the testing data. There are three phases to acquire the rules automatically: (1) segmentation of the target words in which the inflected forms of the verbs are divided into three splits. The first split is the part of target before the pattern key string is found, whereas the second split is considered the base form and the third split is automatically considered the suffix added to the verb; (2) discovering the affix/es added to the verb at the beginning or at the middle of the base form by processing the first split taken from segmentation and (3) acquiring the rule based from the affix/es added to the verbs based from the two succeeding processes expressed in a notation. The algorithm has achieved 89% and 87% accuracy in morphological generation and analysis respectively.

Abstract Format

html

Language

English

Format

Electronic

Accession Number

CDTG004590

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

x, 110 leaves ; 28 cm.

Keywords

Tagalog language--Verb; Natural language processing (Computer science); Morphology (Linguistics)

Upload Full Text

wf_yes

Download

Share

COinS