PiE: Programming in Eliza

Eliza, a primitive example of natural language processing, adopts a rule-based method to conduct simple conversations with people. In this paper, we extend Eliza for a novel application. We pro-pose a system to assist with program synthesis called Programming in Eliza (PiE). According to a set of rules, PiE can automatically synthesize programs from natural language conversations between Eliza and users. PiE is useful for programming in domain-specific languages. We have implemented PiE to synthesize programs in the LOGO programming language, and our experimental results show that, on average, the success ratio is 88.4 % for synthesizing LOGO programs from simple conversations with Eliza. PiE also enables end-users with no experience to program in LOGO with a smoother learning curve