近年行動裝置的硬體技術發展迅速,通訊上速度與頻寬也不斷升級,加上使用者介面的重大突破,進而使的行動運算與行動雲端運算也一同蓬勃發展起來。在軟體的方面,行動裝置應用程式的開發也更為容易與簡便,已經有十幾萬種行動應用程式,但大部分的行動應用很少完整使用行動裝置的高移動性、便利性、個人化及位置化等特性,使用者在使用行動應用時,往往不能貼近於使用者需求而提供相關資訊。因此本研究提出在行動應用導入情境感知的概念,藉此來解決使用者在使用行動應用時,所面臨的資訊非使用者所需之問題。 本研究透過導入情境感知於行動運算系統架構以及作業流程中,以改善資訊的適切性。系統設計方面運用UML的案例圖、時序圖以及活動圖等工具,並設計系統所需的資料庫與演算法。當使用者使用行動應用時,根據使用的功能,應用程式主動提供所需的情境資訊,配合資料庫內的資訊在網路伺服器中進行演算,以提供使用者所需的資訊。 本研究以租屋資訊系統為例,將情境感知的概念導入租屋流程,實作出具情境感知的租屋資訊應用,藉由兩個實作情境來與現今的租屋資訊應用做比較,來確認情境感知改善行動應用的效果。本研究以一般動態網頁的三層式架構為基礎,將情境感知演算法置於網路伺服器中,實作的結果顯示,使用者認為具有情境感知的搜尋可提供更貼切的資訊,進而提升使用者的使用滿意度。
In recent years, mobile device hardware has developed rapidly, communication bandwidth and speed have also upgraded, and there has been a major breakthrough in user interface. Therefore, mobile computing and mobile cloud computing have dramatically prospered. Mobile application development has become easier and more convenient with better software, and there have already been more than ten thousand mobile applications. However, most of the mobile applications are not developed with fully usage of mobile device properties, like high mobility, convenience, personalization and location. This study proposed mobile applications should leverage context awareness to bridge the information gap between user requirements and mobile applications. In system design stage, UML tools like case diagram, sequence diagram and activity diagrams were applied. The database and context-aware algorithms were developed to fulfill the system requirements. The idea is the application would provide search results based on not only the user-specified criteria but also user preference actively derived from the database information. In this study, a house rental listings system was used as an example to show how context awareness was introduced to the rental listings application. Two search scenarios were implemented to verify whether context awareness would improve the mobile applications. The three-tier architecture of dynamic web application was used to implement the listings system, and the context-aware algorithms were realized in the web server. The experimental results show that search activities with context awareness provide more relevant information and get better user satisfactions.