Mining online software tutorials: challenges and open problems

Web-based software tutorials contain a wealth of information describing software tasks and workflows. There is growing interest in mining these resources for task modeling, automation, machine-guided help, interface search, and other applications. As a first step, past work has shown success in extracting individual commands from textual instructions. In this paper, we ask: How much further do we have to go to more fully interpret or automate a tutorial? We take a bottom-up approach, asking what it would take to: (1) interpret individual steps, (2) follow sequences of steps, and (3) locate procedural content in larger texts.