Paper Details: Downloads: 2260
Serial Number: P1121713558
Title: GA-based Parameter Optimization for Word Segmentation
Authors: Ammar Mohammed and Mohammed Karam and Hesham Hefny
Abstract: word segmentation is the process of finding the best likely sequence of words from a sequence of concatenated characters without spaces. Several researches proposed solutions to word segmentations using heuristic methods. The main task of the last methods is to hopefully find the best segmentation without searching the entire state spaces. This paper proposes a new approach for word segmentation based on parameters optimization by means of Genetic Algorithm. The approach is tested on English language using two different language models taking into consideration several data sets. To show that the presented approach is domain language independent, the approach is experimented furthermore on the Arabic language. The experiments show that segmentation using parameters optimization gives a better results.
Keywords: Word segmentation, Genetic Algorithm
Journal/Conference: International Journal of Artificial Intelligence and Machine Learning
Volume: 17
Issue: 1
Submission Date: 3/26/2017 12:00:00 AM
Review Date: 5/3/2017 12:00:00 AM
Publishing Date: 5/23/2017 12:00:00 AM
Article Downloads: 2260
Download:

Facebook