#The cocomo model software
Effort estimates may be used as input to project plans, iteration plans, budgets, and investment analyses, pricing processes and bidding rounds.The use of a repeatable, clearly defined and well understood software development process has, in recent years, shown itself to be the most effective method of gaining useful historical data that can be used for statistical estimation. Software development efforts estimation is the process of predicting the most realistic use of effort required to develop or maintain software based on incomplete, uncertain and/or noisy input. and models which are classified as algorithmic and non-algorithmic approach.
There are many software cost estimation techniques Section 7 ends the paper with a conclusion. Experimental results and evaluation criteria are shown in section 6.
Section 4 discusses the related work and proposed neural network model and its algorithm is described in section 5. The paper is organized in following sections: section 1 describes introduction, sections 2 and 3 describes COCOMO II model and neural network using perceptron learning rule. Many researchers are working in implementing software effort and cost estimation in neural networks. Perceptron model is supervised model of neural network where weights are updated depending on the teachers response. In this paper a neural network technique using perceptron learning algorithm for software cost estimation which is based on COCOMO II model is proposed. The non-algorithm based algorithm work with real life situations and a vast flexibility for software development factors was provided. Due to few such limitations of conventional algorithmic models,non- algorithmicmodels based on Soft Computing came into picture, which include Neural Network, Fuzzy logic and Genetic algorithms. Values of software development factors based on experience and approximation, with zero reasoning Capability. These models require inputs which are difficult to obtain at early stages of software development. One of the most frequently used model to estimate software effort is COCOMO developed by Berry Boehm. The models developed were based on mathematical formula and software development factors. In past few decades several researchers have worked in the field of software effort estimation, and many conventional models were designed to estimate software, size and effort. Since the effort and cost estimation is done at an early stage of software development hence a good model is required to calculate these parameters accurately. For efficient software, accurate software development parameters are required, these include effort estimation, development time estimation, cost estimation, team size estimation, risk analysis, etc. With an effective estimate of software cost and effort, software developers can efficiently decide what recourses are to be used frequently and how efficiently these resources can be utilized. It is the accuracy of cost and effort calculation that enable quality growth of software at a later stage. Software cost and effort estimate is one of the most important activities in software project management. Keywords: COCOMO II, Neural Networks, Perceptron learning rule. This work proposes an estimation model that incorporates COCOMO II with perceptron learning rule to provide more accurate software estimates at early phase of software development, so that the estimated effort is more close to the actual effort. In this paper, the author explores the use of perceptron learning rule to implement COCOMO II for effort estimation. In software industry the most widely used model for effort estimation is Constructive Cost Model (COCOMO). Since it is very difficult to bridge the gap between estimated cost and actual cost, hence the accurate cost estimation is one of the challenging tasks in maintaining software projects. Software cost and effort estimation is the most critical task in handling software projects. Guru Gobind Singh Indraprastha University, New Delhi Rama Kishore2 1M.Tech (IT) Scholar, 2Asstt. COCOMO II Implementation Using Perceptron Learning Rule
0 kommentar(er)
