当前位置: X-MOL 学术Int. J. Softw. Eng. Knowl. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Applicability of Machine Learning Methods on Mobile App Effort Estimation: Validation and Performance Evaluation
International Journal of Software Engineering and Knowledge Engineering ( IF 0.9 ) Pub Date : 2020-02-26 , DOI: 10.1142/s0218194020500023
Mamta Pandey 1 , Ratnesh Litoriya 1 , Prateek Pandey 1
Affiliation  

Software cost estimation is one of the most crucial tasks in a software development life cycle. Some well-proven methods and techniques have been developed for effort estimation in case of classical software. Mobile applications (apps) are different from conventional software by their nature, size and operational environment; therefore, the established estimation models for traditional desktop or web applications may not be suitable for mobile app development. The objective of this paper is to propose a framework for mobile app project estimation. The research methodology adopted in this work is based on selecting different features of mobile apps from the SAMOA dataset. These features are later used as input vectors to the selected machine learning (ML) techniques. The results of this research experiment are measured in mean absolute residual (MAR). The experimental outcomes are then followed by the proposition of a framework to recommend an ML algorithm as the best match for superior effort estimation of a project in question. This framework uses the Mamdani-type fuzzy inference method to address the ambiguities in the decision-making process. The outcome of this work will particularly help mobile app estimators, development professionals, and industry at large to determine the required efforts in the projects accurately.

中文翻译:

机器学习方法在移动应用努力估计中的适用性:验证和性能评估

软件成本估算是软件开发生命周期中最关键的任务之一。在经典软件的情况下,已经开发了一些经过充分验证的方法和技术来估计工作量。移动应用程序(应用程序)在性质、规模和运行环境方面与传统软件不同;因此,为传统桌面或 Web 应用程序建立的估计模型可能不适合移动应用程序开发。本文的目的是提出一个移动应用程序项目估算的框架。这项工作采用的研究方法是基于从 SAMOA 数据集中选择移动应用程序的不同特征。这些特征稍后用作所选机器学习 (ML) 技术的输入向量。该研究实验的结果以平均绝对残差 (MAR) 来衡量。实验结果之后是一个框架的提议,以推荐 ML 算法作为对所讨论项目的卓越工作量估计的最佳匹配。该框架使用 Mamdani 型模糊推理方法来解决决策过程中的歧义。这项工作的结果将特别有助于移动应用程序评估人员、开发专业人员和整个行业准确地确定项目所需的工作量。该框架使用 Mamdani 型模糊推理方法来解决决策过程中的歧义。这项工作的结果将特别有助于移动应用程序评估人员、开发专业人员和整个行业准确地确定项目所需的工作量。该框架使用 Mamdani 型模糊推理方法来解决决策过程中的歧义。这项工作的结果将特别有助于移动应用程序评估人员、开发专业人员和整个行业准确地确定项目所需的工作量。
更新日期:2020-02-26
down
wechat
bug