完整後設資料紀錄
DC 欄位語言
dc.contributor.author張志平zh_TW
dc.contributor.author王豐堅zh_TW
dc.contributor.authorChang, Chih-Pingen_US
dc.contributor.authorWang, Feng-Jianen_US
dc.date.accessioned2018-01-24T07:42:04Z-
dc.date.available2018-01-24T07:42:04Z-
dc.date.issued2016en_US
dc.identifier.urihttp://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070256550en_US
dc.identifier.urihttp://hdl.handle.net/11536/142339-
dc.description.abstract目前在雲端系統上選取一個最佳的服務的研究,大部分根據QoS的預測值來完成。現存的預測方法通常運用一個獨特的計算方式伴隨著一個或多個額外特性來改善預測準確度。然而,目前沒有比較這些方法的研究被提出來。在本論文中,我們針對現有的方法做了一個調查並得到一有效分類。第一,我們介紹現有QoS預測的基本計算方法並且將他們分成以記憶和以模型為基礎計算的類型。第二,為了改善先前的基本方法,我們也調查現存的一些使用者用來改善的準則,並加以分類為相似度類型以及分群類型。由於這兩種分類係正交,因此我們可將現存方法歸納成四大類,並將每一類的方法做對應之比較。另外,我們也更進一步對所有方法做精確、時效與應用範圍的比較。zh_TW
dc.description.abstractCurrent studies for selecting a service are done according to its predicted QoS value(s), and each of existing approaches introduces a distinct prediction method with one or more user selection criterion to improve the accuracy of QoS prediction. However, there are no researches to make a comparison among these approaches. In this thesis, we make a survey on existing approaches. First, the basic methods for QoS prediction can be divided into two models: memory and model-based. Second, the user selection criteria to improve the above basic methods can be classified into two categories: similarity and clustering. The existing approaches categorized based on basic methods and improvement approaches are then discussed separately and a final comparison is shown then.en_US
dc.language.isozh_TWen_US
dc.subject服務導向架構zh_TW
dc.subjectQoS 預測zh_TW
dc.subject概觀調查zh_TW
dc.subject協同過濾zh_TW
dc.subjectService-Oriented Architectureen_US
dc.subjectSurveyen_US
dc.subjectQoS Predictionen_US
dc.subjectCollaborative Filteringen_US
dc.title一個網路服務QoS預測之研究與效能評估zh_TW
dc.titleA Study and Performance Assessment of QoS Prediction for Web Servicesen_US
dc.typeThesisen_US
dc.contributor.department網路工程研究所zh_TW
顯示於類別:畢業論文