Towards computer-generated mnemonic phrases: Experiments with genetic algorithms and N-grams

Mountstephens, James , (2014) Towards computer-generated mnemonic phrases: Experiments with genetic algorithms and N-grams. ARPN Journal of Engineering and Applied Sciences, 9 (9). pp. 1386-1393. ISSN 1819-6608

Full text not available from this repository.

Abstract

Mnemonic phrases have the potential to help people commit information to memory and may be a valuable aid to education. However, their widespread application is currently limited by the effort and creativity required to generate them manually. This paper describes a method for the automatic generation of effective mnemonics by computer, framing the task as an optimisation problem to be solved by Genetic Algorithms using parser output and n-gram frequencies to evaluate fitness. Grammatical constraints and lexical familiarity are parameters tested for their ability to produce more memorable sentences. The method has been implemented using custom code and existing libraries, and tested, showing promising results on list data of increasing difficulty.

Item Type: Article
Keyword: Education; Genetic algorithms; Mnemonics; N-grams; Natural language
Subjects: ?? QA75 ??
T Technology > TA Engineering (General). Civil engineering (General)
Department: FACULTY > Faculty of Computing and Informatics
Depositing User: ADMIN ADMIN
Date Deposited: 23 Feb 2015 10:35
Last Modified: 23 Feb 2015 10:35
URI: https://eprints.ums.edu.my/id/eprint/10533

Actions (login required)

View Item View Item