Atlantis Press Paper Details: String Filtering of a Large String Collection on Mobile Devices using a Neural Network

String matching of a large string collection on mobile devices has been a difficult problem because of the memory space and computing speed constraints. We propose a method to efficiently determine whether a query string exists in the large string collection. The proposed method, based on a string encoder using a neural network, can perform the task in a consistent manner regardless of the number of strings in the collection. The string encoding process is executed on the server site, and the result is transmitted to the PDAs via the Internet. The server checks a remote database containing the most-updated data constantly, and thus renders the PDAs the accuracy in filtering the query strings. We conducted an experiment that simulates the stolen car license plate matching, and the results were favorable in both speed and memory space requirement.

		back to author index

title:		String Filtering of a Large String Collection on Mobile Devices using a Neural Network
publication:		JCIS-2006 Proceedings
part of series:		Advances in Intelligent Systems Research
ISBN:		978-90-78677-01-7
ISSN:		1951-6851
DOI:		doi:10.2991/jcis.2006.120 (how to use a DOI)
author(s):		Heng Ma, Chia-Cheng Liu
corresponding author:		Heng Ma
publication date:		October 2006
keywords:		Text String, Mobile Devices, Neural Network
abstract:		String matching of a large string collection on mobile devices has been a difficult problem because of the memory space and computing speed constraints. We propose a method to efficiently determine whether a query string exists in the large string collection. The proposed method, based on a string encoder using a neural network, can perform the task in a consistent manner regardless of the number of strings in the collection. The string encoding process is executed on the server site, and the result is transmitted to the PDAs via the Internet. The server checks a remote database containing the most-updated data constantly, and thus renders the PDAs the accuracy in filtering the query strings. We conducted an experiment that simulates the stolen car license plate matching, and the results were favorable in both speed and memory space requirement.
copyright:		© Atlantis Press. This article is distributed under the terms of the Creative Commons Attribution License, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited.
full text:		JCIS06-CSI-75.pdf (592 K)