A data structure between trie and list for auto completion

Boo, Vooi Keong and Patricia Anthony (2012) A data structure between trie and list for auto completion. Communications in Computer and Information Science, 295. pp. 303-312. ISSN 1865-0929


Download (42kB) | Preview


Auto completion is one of the useful features available as a web service. This technique can be implemented in many applications, ranging from a small scale of service like auto complete for items sold in a shop to a large scale of service like Google suggestions which involve suggesting a huge dataset. One of the challenges in implementing auto complete service with a large dataset and a limited computer power is on how to achieve a fast lookup without consuming a lot of memory. This paper presents a data structure that can implement auto complete service that contains up to millions of concepts in a standard computer. The proposed data structure increases the search complexity in return to saving a large amount of memory. A service similar to DBpedia lookup service which contains 9 million words as completion candidates is developed to test the performance of this data structure. The testing shows that such data structure requires less memory than ternary search tree and more importantly a lookup can be performed within milliseconds.

Item Type: Article
Keyword: Auto completion, Search tree, trie, Auto completion, Computer power, Data sets, Lookup services, Lookups, Search complexity, Search trees, Small scale, trie, Forestry, Web services, Data structures, Data, Forestry, Structures
Subjects: Q Science > QA Mathematics > QA1-939 Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
?? Z665-718.8 ??
Department: SCHOOL > School of Engineering and Information Technology
Depositing User: ADMIN ADMIN
Date Deposited: 14 Nov 2012 11:06
Last Modified: 20 Oct 2017 14:58
URI: https://eprints.ums.edu.my/id/eprint/5383

Actions (login required)

View Item View Item