A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing
Recommender systems are information filtering systems that cope with the issue of excessive data by filtering fragments of important information. The massive amount of information is dynamically generated according to the user’s preferences, interests, or observed behaviour of an item. Recommender s...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English English |
Published: |
2021
|
Subjects: | |
Online Access: | http://eprints.utem.edu.my/id/eprint/25975/1/A%20multi-criteria%20hybrid%20recommender%20system%20for%20elderly%20intervention%20plans%20toward%20successful%20ageing.pdf http://eprints.utem.edu.my/id/eprint/25975/2/A%20multi-criteria%20hybrid%20recommender%20system%20for%20elderly%20intervention%20plans%20toward%20successful%20ageing.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-utem-ep.25975 |
---|---|
record_format |
uketd_dc |
spelling |
my-utem-ep.259752022-09-29T12:33:19Z A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing 2021 Azmi, Aini Khairani T Technology (General) TK Electrical engineering. Electronics Nuclear engineering Recommender systems are information filtering systems that cope with the issue of excessive data by filtering fragments of important information. The massive amount of information is dynamically generated according to the user’s preferences, interests, or observed behaviour of an item. Recommender systems have been widely applied in many domains, such as e-commerce, health, food, and nutrition, movies, and many others. Currently, numerous endeavors have been made to improve the lives of those people who are elderly using recommender systems. Current assessment only focusing on a single assessment process which not comprehensive. The assessment is often used to determine the current and future interventions that should be given accurately to the elderly. To ensure that intervention plans are provided comprehensively to the elderly, many aspects need to be addressed. This research proposes a hybrid recommender system that combines both collaborative filtering (CF) and knowledge-based (KB) approaches based on the profiles of elderly people generated from the elderly assessments. The user profile represents the elderly condition for each aspect of assessments and will be used by the proposed model to recommend interventions for the elderly. The CF was applied for determining similar users based on the profiles of other users. The KB filtering technique was then applied to select the interventions listed by the CF approach based on the interventions given by the experts who participated in this research to improve the well-being of the elderly and helping them to achieve successful ageing. The proposed recommendation model was evaluated based on its accuracy by using precision, recall, and F1 Measure to compare the proposed model with the baseline models using basic search (BS) and CF to determine which recommendation model was preferred in recommending interventions based on multi aspects of successful ageing. The result from the accuracy evaluation using recall, precision, and F1 Measure revealed that the new recommendation model that integrates both CF and KB approach are more accurate compare to baseline model. The Successful Ageing Method (SAM) system that has been developed in this research using this new recommendation model can be used for the elderly institutions under JKM aligned with the Rancangan Malaysia Ke-12 (RMK-12). This will helps the elderly care sector to be better in organizing and taking care of elderly well-being. 2021 Thesis http://eprints.utem.edu.my/id/eprint/25975/ http://eprints.utem.edu.my/id/eprint/25975/1/A%20multi-criteria%20hybrid%20recommender%20system%20for%20elderly%20intervention%20plans%20toward%20successful%20ageing.pdf text en public http://eprints.utem.edu.my/id/eprint/25975/2/A%20multi-criteria%20hybrid%20recommender%20system%20for%20elderly%20intervention%20plans%20toward%20successful%20ageing.pdf text en validuser https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=121050 mphil masters Universiti Teknikal Malaysia Melaka Faculty of Information and Communication Technology Abdullah, Noraswaliza |
institution |
Universiti Teknikal Malaysia Melaka |
collection |
UTeM Repository |
language |
English English |
advisor |
Abdullah, Noraswaliza |
topic |
T Technology (General) T Technology (General) |
spellingShingle |
T Technology (General) T Technology (General) Azmi, Aini Khairani A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
description |
Recommender systems are information filtering systems that cope with the issue of excessive data by filtering fragments of important information. The massive amount of information is dynamically generated according to the user’s preferences, interests, or observed behaviour of an item. Recommender systems have been widely applied in many domains, such as e-commerce, health, food, and nutrition, movies, and many others. Currently, numerous endeavors have been made to improve the lives of those people who are elderly using recommender systems. Current assessment only focusing on a single assessment process which not comprehensive. The assessment is often used to determine the current and future interventions that should be given accurately to the elderly. To ensure that intervention plans are provided comprehensively to the elderly, many aspects need to be addressed. This research proposes a hybrid recommender system that combines both collaborative filtering (CF) and knowledge-based (KB) approaches based on the profiles of elderly people generated from the elderly assessments. The user profile represents the elderly condition for each aspect of assessments and will be used by the proposed model to recommend interventions for the elderly. The CF was applied for determining similar users based on the profiles of other users. The KB filtering technique was then applied to select the interventions listed by the CF approach based on the interventions given by the experts who participated in this research to improve the well-being of the elderly and helping them to achieve successful ageing. The proposed recommendation model was evaluated based on its accuracy by using precision, recall, and F1 Measure to compare the proposed model with the baseline models using basic search (BS) and CF to determine which recommendation model was preferred in recommending interventions based on multi aspects of successful ageing. The result from the accuracy evaluation using recall, precision, and F1 Measure revealed that the new recommendation model that integrates both CF and KB approach are more accurate compare to baseline model. The Successful Ageing Method (SAM) system that has been developed in this research using this new recommendation model can be used for the elderly institutions under JKM aligned with the Rancangan Malaysia Ke-12 (RMK-12). This will helps the elderly care sector to be better in organizing and taking care of elderly well-being. |
format |
Thesis |
qualification_name |
Master of Philosophy (M.Phil.) |
qualification_level |
Master's degree |
author |
Azmi, Aini Khairani |
author_facet |
Azmi, Aini Khairani |
author_sort |
Azmi, Aini Khairani |
title |
A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
title_short |
A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
title_full |
A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
title_fullStr |
A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
title_full_unstemmed |
A multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
title_sort |
multi-criteria hybrid recommender system for elderly intervention plans toward successful ageing |
granting_institution |
Universiti Teknikal Malaysia Melaka |
granting_department |
Faculty of Information and Communication Technology |
publishDate |
2021 |
url |
http://eprints.utem.edu.my/id/eprint/25975/1/A%20multi-criteria%20hybrid%20recommender%20system%20for%20elderly%20intervention%20plans%20toward%20successful%20ageing.pdf http://eprints.utem.edu.my/id/eprint/25975/2/A%20multi-criteria%20hybrid%20recommender%20system%20for%20elderly%20intervention%20plans%20toward%20successful%20ageing.pdf |
_version_ |
1747834146252652544 |