ABSTRACT
IT companies are embracing Agile methodologies for their current and upcoming software projects. The most challenging aspect is project management. Many CASE tools came up as solution to handle dynamic specificities of Scrum based projects. The present state reveals companies rely on the traditional ways of handling estimation issues of software projects which suffers from individual bias and subjective assessment. Scrum projects are volatile and as per the Agile Manifesto incur changes in the various phases. Story point is a relative metric of measuring effort of desired user stories in Scrum. It has been classified and predicted using traditional and machine learning techniques. In this paper, we have proposed a novel framework for story point classification in Scrum projects using deep learning. The proposed framework is end-to-end trainable and lays the foundation of a continuous estimation framework.
