In the software measurement validations, assessing the validation of software metrics in software engineering is a very difficult task due to lack of theoretical methodology and empirical methodology [41, 44, 45]. During recent years, there have been a number of researchers addressing the issue of validating software metrics. At present, software metrics are validated theoretically using properties of measures. Further, software measurement plays an important role in understanding and controlling software development practices and products. The major requirement in software measurement is that the measures must represent accurately those attributes they purport to quantify and validation is critical to the success of software measurement. Normally, validation is a collection of analysis and testing activities across the full life cycle and complements the efforts of other quality engineering functions and validation is a critical task in any engineering project. Further, validation objective is to discover defects in a system and assess whether or not the system is useful and usable in operational situation. In the case of software engineering, validation is one of the software engineering disciplines that help build quality into software. The major objective of software validation process is to determine that the software performs its intended functions correctly and provides information about its quality and reliability. This paper discusses the validation methodology, techniques and different properties of measures that are used for software metrics validation. In most cases, theoretical and empirical validations are conducted for software metrics validations in software engineering [1-50].

This paper introduces a procedure that has been developed for evaluating an object-oriented design of a system that involves many classes. This approach involves two new metrics called Total Class Metric (TCM) and Total System Metric (TSM) that assess the design of a class and system as a whole respectively during object-oriented development process. In the increasing use of object-orientation in software development, there is a growing need to measure efficiency and effectiveness of the design process. In response to this need, a number of researchers have developed various metrics for object-oriented systems. A procedure has been introduced for evaluating the effectiveness of the object-oriented design of a system for the improvement of the software process instead of using individual design metrics. The total class metric is defined based on a set of seven metrics which have been formulated using main attributes and significant characteristics of an object-oriented design of the system. This research paper discusses in detail about the new approach, total class metric and total system metric to represent the single quality value for the entire system design to judge the effectiveness of the design. These metrics will be useful in measuring object-oriented design and feedback system of software measurement thus yielding an effective object-oriented design.

The software reusability mode is highly required field for successful execution of artificial intelligence, machine learning based applications to fulfill the present and future human needs. The identification, classification and measuring the required components are key-roles concerns in fast development of software reusability components for producing the high quality software. This paper is proposing the Fuzzy Logic Controller Neural Network Hybrid System which is implicated to recognize the affecting factors of component reusability execution by instituting the strong, week relationships in among these considered factors to fulfill the user requirement. This approach considered eleven effecting factors such as Portability, Reliability, Complexity, Efficiency, Quality, Security, Cost, Maintainability, Cohesion, Availability and Flexibility along with their related attribute metrics. This paper has composed with four major objectives such as: the comparative analysis of Fuzzy Logic Control System and Neural Networks with their advantages and execution flow; The implications of Fuzzy Logic Control Neural Network Hybrid System architecture design for concern problem; The proposed FLCNNHS based algorithm and execution data flow diagram for executing the considered software reusability effecting factors along with their supporting attributes metrics for identification and Classification of Reusability Components through Strong, Week Relationships of Lattice Factors which is implacable for designing the better quality software product; and described the experimental analysis and results through proposed algorithmic approach. This innovative approach is more helpful for software developers to choose highly accurate components which are more required to build the high efficiency secure systems.

<span lang="EN-US">In component based software reusability development process, the software developers have to choose the best components which are self adaptive future to overcome the functional errors, framework mismatches, violation of user level privacy issues and data leakage feasibilities. The software developers can build high quality software applications by taking the consideration of the reusable components which are more suitable to provide high level data security and privacy. This paper has proposing the neural based fuzzy framework based approach to estimate the reusable components which are directly and indirectly involve the security and privacy to improve the quality of the software system. This approach has considered the twenty effecting factors and fifty three attribute matrices. It has formed with three stages of execution scenarios. The first stage has executed with eleven effecting factors and eighteen attribute matrices for identification of supporting software reusability components, the second stage has executed with four effecting factors and thirty five attribute matrices for identification of sub-internal relationships in terms of security-privacy, and the third stage has executed with eight effecting factors and six attribute matrices for identification of sub of sub-internal relationships in terms of security risk estimation. This analytical finding proposes a fuzzy logic model to evaluate the most feasible effecting factors that influence the enterprise level data security-privacy practices at real time environment.</span>

No abstract