東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Three-phase Decision Making for Self...

Al Kafaf, Dhrgam.

FindBook

Google Book

Amazon

博客來

Three-phase Decision Making for Self-adaptive Systems.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Three-phase Decision Making for Self-adaptive Systems./
作者:	Al Kafaf, Dhrgam.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
面頁冊數:	138 p.
附註:	Source: Dissertations Abstracts International, Volume: 80-01, Section: B.
Contained By:	Dissertations Abstracts International80-01B.
標題:	Computer Engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10751855
ISBN:	9780438184664

Three-phase Decision Making for Self-adaptive Systems.
Al Kafaf, Dhrgam.

Three-phase Decision Making for Self-adaptive Systems. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 138 p.

Source: Dissertations Abstracts International, Volume: 80-01, Section: B.

Thesis (Ph.D.)--Oakland University, 2018.

This item must not be sold to any third party vendors.

A self-adaptive system adapts itself to changes in a dynamic environment. The core in self-adaptive systems is making an adaptation decision. The current practice focuses on a single layer of decision making using either a local knowledge base or a shared knowledge base by multiple systems through a network. While the use of a local knowledge base is efficient, it suffers from the limited maturity of the knowledge base due to the sole learning. A shared knowledge base can address the maturity problem, but it is inefficient in adaptation due to communication overheads. This research presents a three-phase decision making approach for self-adaptive systems to improve precision while being competitive for efficiency. The approach consists of three phases for making a decision. The first phase uses the local knowledge base of the self-adaptive unit for an object that can be identified locally. If the object cannot be identified locally, the unit sends a request to shared knowledge bases through Web services in the second and third phases. Web services are used for accessing shared knowledge bases. The approach presented in this research uses machine learning techniques for decision making based on the kNN algorithm. The kNN algorithm is chosen in this research due to its simplicity and high precision rate compared to other algorithms. However, the kNN algorithm is not efficient with regards to processing time. This led this research to focus on improving the performance of the kNN algorithm. The kNN algorithm relies on the exhaustive use of training datasets, which reduces efficiency on large datasets. This research presents the B-kNN algorithm which improves the efficiency of kNN using a two-fold preprocess scheme built upon the notion of minimum and maximum points and boundary subsets. For a given training dataset, B-kNN first identifies classes, and for each class, it further identifies the minimum and maximum points (MMP) of the class. A given testing object is evaluated to the MMP of each class. If the object belongs to the MMP, the object is predicted to belong to the class. If not, a boundary subset (BS) is denied for each class. Then, BSs are fed into kNN to determine the class of the object. As BSs are significantly smaller in size than their classes, the efficiency of kNN improves. The approach in this research is validated through implementation using Gazebo, a 3D simulation environment, and applying the implementation to robotic systems. The validation uses five different scenarios. The results show 96% precision in identifying objects with a viable overhead introduced by the web service and 45% improvement in precision over the traditional approach. This research analyzes the outcomes in terms of precision and response time.

ISBN: 9780438184664Subjects--Topical Terms:

1567821
Computer Engineering.

Three-phase Decision Making for Self-adaptive Systems.
LDR:03830nmm a2200325 4500 001 2208073
005 20190929184211.5
008 201008s2018 ||||||||||||||||| ||eng d
020 $a 9780438184664
035 $a (MiAaPQ)AAI10751855
035 $a (MiAaPQ)oakland:10075
035 $a AAI10751855
040 $a MiAaPQ $c MiAaPQ
100 1 $a Al Kafaf, Dhrgam. $3 3435085
245 1 0 $a Three-phase Decision Making for Self-adaptive Systems.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 138 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-01, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Kim, Dae-Kyoo.
502 $a Thesis (Ph.D.)--Oakland University, 2018.
506 $a This item must not be sold to any third party vendors.
520 $a A self-adaptive system adapts itself to changes in a dynamic environment. The core in self-adaptive systems is making an adaptation decision. The current practice focuses on a single layer of decision making using either a local knowledge base or a shared knowledge base by multiple systems through a network. While the use of a local knowledge base is efficient, it suffers from the limited maturity of the knowledge base due to the sole learning. A shared knowledge base can address the maturity problem, but it is inefficient in adaptation due to communication overheads. This research presents a three-phase decision making approach for self-adaptive systems to improve precision while being competitive for efficiency. The approach consists of three phases for making a decision. The first phase uses the local knowledge base of the self-adaptive unit for an object that can be identified locally. If the object cannot be identified locally, the unit sends a request to shared knowledge bases through Web services in the second and third phases. Web services are used for accessing shared knowledge bases. The approach presented in this research uses machine learning techniques for decision making based on the kNN algorithm. The kNN algorithm is chosen in this research due to its simplicity and high precision rate compared to other algorithms. However, the kNN algorithm is not efficient with regards to processing time. This led this research to focus on improving the performance of the kNN algorithm. The kNN algorithm relies on the exhaustive use of training datasets, which reduces efficiency on large datasets. This research presents the B-kNN algorithm which improves the efficiency of kNN using a two-fold preprocess scheme built upon the notion of minimum and maximum points and boundary subsets. For a given training dataset, B-kNN first identifies classes, and for each class, it further identifies the minimum and maximum points (MMP) of the class. A given testing object is evaluated to the MMP of each class. If the object belongs to the MMP, the object is predicted to belong to the class. If not, a boundary subset (BS) is denied for each class. Then, BSs are fed into kNN to determine the class of the object. As BSs are significantly smaller in size than their classes, the efficiency of kNN improves. The approach in this research is validated through implementation using Gazebo, a 3D simulation environment, and applying the implementation to robotic systems. The validation uses five different scenarios. The results show 96% precision in identifying objects with a viable overhead introduced by the web service and 45% improvement in precision over the traditional approach. This research analyzes the outcomes in terms of precision and response time.
590 $a School code: 0446.
650 4 $a Computer Engineering. $3 1567821
650 4 $a Computer science. $3 523869
690 $a 0464
690 $a 0984
710 2 $a Oakland University. $b Engineering. $3 3288992
773 0 $t Dissertations Abstracts International $g 80-01B.
790 $a 0446
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10751855