Scholarlink

<p>Software effort estimation is a critical and vital step in the software development life cycle. Software effort estimation has attracted significant attention from both the software industry and the research community. Effort estimates are crucial for determining development time, resource planning, and budgeting. Therefore, accurately estimating effort is essential for effective project management. Numerous researchers have previously presented a variety of techniques for estimating software effort, including statistical, algorithmic, machine learning-based, and nature-inspired models. Selecting the most accurate estimation method is a complex procedure. This study optimizes the COCOMO-II model parameters for estimation accuracy by employing the Firefly Algorithm(FA) and the Chaotic Firefly Algorithm with Enhanced Exploration (CFAEE). Mean Magnitude of Relative Error (MMRE) and Prediction (PRED) are used to assess the application of the two models on six datasets. While the Firefly Algorithm surpasses previous algorithms, the results show that the CFAEE methodology is the most effective approach for determining the efficiency coefficients of the COCOMO-II model. Statistical testing also shows that the proposed model yields better results.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Software development effort estimation using COCOMO-II model with FA and CFAEE

  • 期刊论文

  • 发表日期：

    2025年11月26日

查看全文
  • Sarika Mustyala,
  • Manjubala Bisi

摘要

Software effort estimation is a critical and vital step in the software development life cycle. Software effort estimation has attracted significant attention from both the software industry and the research community. Effort estimates are crucial for determining development time, resource planning, and budgeting. Therefore, accurately estimating effort is essential for effective project management. Numerous researchers have previously presented a variety of techniques for estimating software effort, including statistical, algorithmic, machine learning-based, and nature-inspired models. Selecting the most accurate estimation method is a complex procedure. This study optimizes the COCOMO-II model parameters for estimation accuracy by employing the Firefly Algorithm(FA) and the Chaotic Firefly Algorithm with Enhanced Exploration (CFAEE). Mean Magnitude of Relative Error (MMRE) and Prediction (PRED) are used to assess the application of the two models on six datasets. While the Firefly Algorithm surpasses previous algorithms, the results show that the CFAEE methodology is the most effective approach for determining the efficiency coefficients of the COCOMO-II model. Statistical testing also shows that the proposed model yields better results.