Poster: Inferring API Elements Relevant to an English Query

We develop a context-sensitive machine translation approach that takes as query a given English description of a programming task, and infers the corresponding API elements for the task by learning from large training corpus. To determine those API elements, we consider the context of the words in the textual query and the context of API elements that often go together in the corpus. We formulate the problem as a set-based statistical machine translation. Our empirical evaluation shows a promising result for our technique.