Towards Automated Generation of User-Specific Eclipse Wizard

Eclipse automates programming-related tasks through wizards. However, as creating a wizard requires a lot of work, the automation is limited on common tasks. In this paper, we propose an automated approach to wizard generation by recording and mining user actions. We record the actions of users in background and learn the common steps of performing tasks by machine learning approach. After that, we synthesize the mined information to generate a wizard automatically. In this way we can generate wizards not only for common tasks but also user-specific tasks. We have implemented our approach as an Eclipse plugin and have evaluated its effectiveness by comparing with existing Eclipse wizards.