NU Science Journal 2006; 2(2): 139 - 150 Hybrid Name Matching Methods for Rule Based Thai Naming System Chakkrit Snae* and Michael Brueckner Department of Computer Science, Faculty of Science, Naresuan University, Phitsanulok 65000 Thailand *Corresponding author. E-mail: chakkrits@nu.ac.th ABSTRACT Names are important in many societies, even in technologically oriented ones which use ID systems or other ways to identify individual people. There are many elements of personal names which vary in and between the different cultures. Names such as personal surnames are the most important as they are used in many processes, such as identifying of people, record linkage and for genealogical research as well. On the other hand, variation of names can be a major problem for the identification and search for people, e.g. web search or security reasons. We show name variations for different cultures to guide the implementation of a rule based naming system, currently worked out for Thai names. We discuss characteristics of the LIG (Levenshtein, Index of Similarity Group (called ISG), and Guth) algorithms which help to find reasonable variants of names. A further benefit of this process would be an optimization for name searching. Keywords: personal names, name variations, name matching, rule based system INTRODUCTION Names are used for identifying persons, places, things and even ideas or concepts. Names serve for labelling of categories or classes and for individual items. They are properties of individuals which are of greater importance in most communities. In technological oriented societies such as modern Western the reference between names as a label and the person is not as obvious as in small tribal societies. This is especially true where names are stored within large information systems. This includes government, medical, educational and even commercial records which are kept about individuals. Names are the most important referrer to a person even if there are numbering systems like ID numbers because such systems are not universal. Names are often queried in a different way than they were entered. Personal names lead to many problems with regard to data retrieval because names are also subject to multiple variations not only between different cultures and writing system but in a specific culture as well. Names represent complex lexical structures which have to be handled systematically for data entry, storage and retrieval in order to get sufficient recall or precision the retrieval process in. In this paper we present a first account of our findings on elements of personal names in different cultures with their respective conventions and variations, rules for constructing Thai names, and algorithms of name matching to overcome name variations. Here we present a hybrid name matching procedure which is based on