標題: 物體導向軟體的複雜性量測之驗證
A Validation of Software Complexity Metrics for Object Oriented Programs
作者: 劉昌鵬
Chang Peng Liu
王豐堅
Feng Jian Wang
資訊科學與工程研究所
關鍵字: 複雜性量測;實驗架構;物體導向程式方法;統計分析模式;驗證準則;資訊流;complexity metric;experiment framework;oop; statistical analysis model;validity criteria;infor. flow
公開日期: 1993
摘要: 物體導向程式方法正迅速被使用在軟體發展領域裡,此技術宣稱具有高度 的軟體再利用性及可維護性。複雜性一般被用來估算程式發展所需耗費的 資源和預測程式的維護性及可靠度。此論文提出一個實驗程序來進行對物 體導向程式的一組複雜性量測的驗證,其中包括我們已發展的一組以資訊 流為基的複雜性量測。在驗證過程中我們採用四個準則,並搭配相關的統 計模式以分析資料。實驗的目標語言是C++,而對象即為用C++語言 所編寫的方法和類別之複雜性及整個發展過程的成本。結果顯示對程式中 的方法及類別,我們提出的量測裡有三個滿足所採用的準則之特性。對方 法及類別發展所需的時間、產生的錯誤和修改的次數,這些驗證過的量測 也可用為程式發展的指標。 Object-oriented programming (OOP) is a new programming concept and technique entering the mainstream of software development rapidly. Claimed advantages of OOP include higher reusability and maintainability. Complexity was commonly used as an attribute of a program. This thesis presents a metrics validation process in which an experimental project was analyzed using sixteen software metrics. The sixteen metrics include information-flow based metrics we have proposed, measures of structure, and measures of size. The validation process used four validity criteria, each has its own statistical analysis model. The experiment was conducted to calculate the complexity and development cost of methods and classes in C++ languages. The results show that three complexity metrics, based on information flow, are valid with four validity criteria. These metrics also work as good indicators of the effort, number of errors, and number of changes of the entity during programming development.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT820392037
http://hdl.handle.net/11536/57842
Appears in Collections:Thesis