Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/441
Title: Analysis of CK and Mood Metrics for Inconsistencies and Preiction of Quality Attributes
Authors: Saini, Sunint
Supervisor: Salaria, R.S.
Keywords: Software engineering;Software Metrics Technology;CK and MOOD Metrics;Computer Science
Issue Date: 14-Nov-2007
Abstract: The need to improve software productivity and software quality has put forward the research on software metrics technology. There is an increasing need for metrics adapted to the Object-Oriented (00) paradigm to help manage and foster quality in software development. Various object-oriented metrics have been proposed by various researchers. Two of the widely accepted metrics are CK and MOOD Metrics. CK and MOOD Metrics have been analyzed according to their validation criteria and it has been observed that CK suite which was build on the validation criteria given by Weyukar fail to satisfy it completely. MOOD metrics on the other hand fail to satisfy the validation criteria given by the MOOD team itself thus showing that MOOD Metrics is working with an inaccurate and imprecise understanding of the 00 paradigm. Hence howing that the genesis of the metrics is controversial. The further inconsistencies in the CK and MOOD Metrics are discussed in detail and it has been observed that none of the metric is without an inconsistency. CK Suite lacks metrics for encapsulation and polymorphism to make it a complete set to measure object-oriented characteristics, while MOOD lacks the metrics for reuse. Some of the existing solutions to the metrics have been discussed and formulae to remove the inconsistencies in Depth of Inheritance Tree and Number of Children have been proposed in this thesis work. The results of the proposed metrics are then compared with the existing metrics as given by CK and it has shown better results as what DIT was originally intended to measure. The Main aim of Object-Oriented metrics is to predict the quality attributes early in the design phase so as to predict the defects and failures and hence reduce the rework and number of dropped modules. Taking this aspect into consideration prediction system has been proposed using fuzzy logic that predicts the fault density using CK metrics and Defect Density using MOOD Metrics. The input to the prediction model that measures fault proneness is WMC, RFC and CBO because these are the metrics which are responsible for measuring fault proneness in the system. The input to the prediction model that predicts the Defect Density is all the six metrics of the MOOD suite. This prediction model predicts the quality attributes early in the design phase and hence speeds up and greatly improves the software development process.
URI: http://hdl.handle.net/123456789/441
Appears in Collections:Masters Theses@CSED

Files in This Item:
File Description SizeFormat 
441.pdf12.46 MBAdobe PDFView/Open    Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.