Software Quality Evaluation by Cocomo II With NN and SVM

Cost, time and quality projection are the crucial aspects in software development process. Incorrect estimations can cause losses which in turn may lead to irreversible damage. It is generally perceived that a imperfectly estimated project always results in a substandard quality due to the efforts being wrongly directed. Firstly Effort Estimation is calculated by actual effort and proposed Effort. That Effort evaluation of 500 NASA projects, after that evaluation is done by four parameters Standard Error, Standard Deviation, Mean Absolute Error, Root Mean Square Error. The author amalgamated the robustness of COCOMO-II with that of Neural Network NN and Support Vector Machine SVM .Quality Which we evaluate that is quality Evaluation of Semantic Web Application. In the last checks the majority of all four parameters for software quality assessment.

Flower Pollination Algorithm for Software Effort Coefficients Optimization to Improve Effort Estimation Accuracy

Software effort estimation is one of important area in project management which used to predict effort for each person to develop an application. Besides, Constructive Cost Model (COCOMO) II is a common model used to estimate effort estimation. There are two coefficients in estimating effort of COCOMO II which highly affect the estimation accuracy. Several methods have been conducted to estimate those coefficients which can predict a closer value between actual effort and predicted value. In this paper, a new metaheuristic algorithm which is known as Flower Pollination Algorithm (FPA) is proposed in several scenario of iteration. Besides, FPA is also compared to several metaheuristic algorithm, namely Cuckoo Search Algorithm and Particle Swarm Optimization. After evaluated by using Mean Magnitude of Relative Error (MMRE), experimental results show that FPA obtains the best result in estimating effort compared to other algorithms by reached 52.48% of MMRE in 500 iterations.

Estimasi Pada Effort Perangkat Lunak dengan Pendekatan Feed Forward Neural Network Backpropagation (FFNN-BP)

Estimasi biaya pengembangan proyek perangkat lunak merupakan salah satu masalah yang kritis dalam rekayasa perangkat lunak. Kegagalan dari proyek perangkat lunak diakibatkan ketidak akuratannya estimasi sumber daya yang dibutuhkan. Beberapa model telah dikembangkan dalam beberapa puluh tahun belakangan ini. Untuk meberikan keakuratan dalam estimasi biaya proyek perangkat lunak masih menjadi tantangan hingga saat ini. Tujuan dilakukannya penelitian ini meningkatkan akurasi estimasi biaya proyek perangkat lunak dengan menerapkan algoritma genetika sebagai proses pelatihan pada Feed Forward Neural Network Backpropagation (FFNN-BP) yang mengakomodasi formula dari Post Architecture Model (COCOMO II). Magnitude of Relative Error (MRE) dan Mean Magnitude of Relative-Error (MMRE) digunakan sebagai pengkuran indikasi kinerja. Hasil percobaan menunjukkan bahwa model yang diusulkan memberikan hasil estimasi biaya proyek perangkat lunak menjadi lebih akurat dari COCOMO II dan FFNN-BP. Dalam kasus ini MMRE untuk COCOMO II adalah 74.68%, FFNN-BP adalah 39.90% . Kata kunci: COCOMO II, Machine Learning, Proyek Manajemen IT, Backpropagation

Refactoring Cost Estimation for Architectural Technical Debt

Paying-off the Architectural Technical Debt by refactoring the flawed code is important to control the debt and to keep it as low as possible. Project Managers tend to delay paying off this debt because they face difficulties in comparing the cost of the refactoring against the benefits gained. These managers need to estimate the cost and the efforts required to conduct these refactoring activities as well as to decide which flaws have higher priority to be refactored. Our research is based on a dataset used by other researchers that study the technical debt. It includes more than 18,000 refactoring operations performed on 33 apache java projects. We applied the COCOMO II:2000 model to calculate the refactoring cost in person-months units per release. Furthermore, we investigated the correlation between the refactoring efforts and two static code metrics of the refactored code. The research revealed a weak correlation between the refactoring efforts and the size of the project, and a moderate correlation with the code complexity. Finally, we applied the DesigniteJava tool to verify our research results. From the analysis we found a significant correlation between the ranking of the architecture smells and the ranking of refactoring efforts for each package. Using machine learning practices, we took the architecture smells level and the code metrics of each release as an input to predict the levels of the refactoring effort of the next release. We calculated the results using our model and found that we can predict the ‘High’ and ‘Very High’ levels, the most significant levels from managers’ perspective, with [Formula: see text] accuracy.

Method for Improving the Efficiency of IT Project Management Using a Genetic Algorithm

The article is devoted to solving the current problem of improving the efficiency of IT project management processes. When managing IT projects, managers are faced with the problem of formation teams and distributing tasks among project participants in the face of the need to minimize costs and completion dates of an IT project. The lack of necessary methods and software doesn't allow the IT project Manager to adequately assess competences and skills of participants, their personal qualities, which leads to a decrease in the effectiveness of project management. The article proposes a method of improving the efficiency of IT project management, which differs by using a genetic algorithm to form project commands and assign team participants to project tasks. The efficiency criterion is the complexity and duration of the project and individual tasks using the COCOMO II method. When forming project teams, takes into account the level of technologies proficiency, experience with technologies, the coherence of the project team members, and the experience of similar developments of project participants. The level of technologies proficiency affects the level of labor input multiplier, experience with technologies — at the level of the multiplier, the coherence of the project team members — on the level of scale factor, the experience of similar development — on the level of the scale factors of the COCOMO II methodology. Taking into account the personal and psychological qualities of project participants reduces the risk of interpersonal conflicts within the team, which also reduces the duration of projects and the labor input of solving tasks. Research of personal and psychological qualities is carried out on the basis of automated tests. The test suite includes Rosenzweig, Belbin, Myers-Briggs, Thomas and Honey-Mumford tests. The developed method is implemented in a software complex for multilevel IT project management. Testing of the method and software complex was carried out within the framework of the students' learning practice of the specialty "Software engineering" of the Belarusian-Russian University. The use of the proposed method allowed to reduce the labor input of solving the tasks of training projects by 19.2 %, to reduce the project realization term by 10 %.

Optimization of software cost estimation model based on biogeography-based optimization algorithm

Estimating of software cost (ESC) is considered a crucial task in the software management life cycle as well as time and quality. Prior to the development of a software project, precise estimations are required in the form of person month and time. In the last few decades, various parametric and non-algorithmic or non-parametric regarding the estimating of software costs have been developed. Among them, the constrictive cost model (COCOMO-II) is a commonly used method for estimating software cost. To further improve the accuracy of this model, researchers and practitioners have applied numerous computational intelligence algorithms to optimize their parameters. However, accuracy is still a big problem in this model to be addressed. In this paper, we proposed a biogeography-based optimization (BBO) method to optimize the current coefficients of COCOMO-II for better estimating of software project cost or effort. The experiments are conducted on two standard data sets: NASA-93 and Turkish Industry software projects. The performance of the proposed algorithm called BBO-COCOMO-II is evaluated by using performance indicators including the Manhattan distance (MD) and the mean magnitude of relative error (MMRE). Simulation results reveal that the proposed algorithm obtained high accuracy and significant error minimization compared to original COCOMO-II, particle swarm optimization, genetic algorithm, flower pollination algorithm, and other various baseline cost estimation models.