Abstract:
Software effort estimation has long been an important task for better software management. Most of the constructed effort estimation models were based on data collected f...Show MoreMetadata
Abstract:
Software effort estimation has long been an important task for better software management. Most of the constructed effort estimation models were based on data collected from software projects that had been developed using traditional software development processes. The structure of this data is usually in the form of tabulated data. Recently, the Agile management framework invaded software production lines as an effective and productive management method. It helps software development teams to complete their tasks in a highly effective way. One of the main components of this success is predicting accurate story points from textual user stories. User stories and story points have become an essential component over which the project planning process is built. Machine learning, artificial intelligence, and deep learning are used to enhance the process of using user story context to put a close estimate of the required resources to finish the project. Using these models has become popular and remarkable in the field of effort estimation. The textual nature of user stories directed the research into the natural language processing path. Natural language processing models can be used to understand textual user story context in order to produce effort estimates. This study reviews the usage of natural language processing methodology in the context of Agile project effort estimation based on the contextual content of user stories.
Date of Conference: 21-23 November 2023
Date Added to IEEE Xplore: 04 December 2023
ISBN Information: