Implementation Of Machine Learning Based Task Allocation In Distributed Agile Software Development

  • Madan Singh, Naresh Chauhan, Rashmi Popli


Agile software development (ASD), has become one of the main stream methodologies for software development in twenty first century. Almost all major multi – national companies have made a switch towards ASD. In the hunt for larger markets and cheaper labor industry has shifted to distributed Agile Software Development (DASD) environment. Task Allocation in distributed environment has been a thrust area and a vital problem as improper allocation of tasks may lead to customer disapproval to accept the project, demonization among team members and further project failures. Numerous researchers have worked on different mechanisms for task allocation in Distributed Agile environment over the past decade. They have applied various techniques like – genetic algorithms, analytic hierarchy process (AHP), domain ontology, Bayesian networks etc. All these techniques are brute force mechanisms that may suit in some specific environment. Further these techniques have not been applied to task allocation in distributed Agile software development.  In this paper an approach for machine learning based task allocation in distributed Agile software development has been proposed and implemented. The results show the proposed model is better in terms of accuracy of task allocation.