Infer XPath

Abstract: We propose reformulation of discovery of data structure within a web page as relations between sets of document nodes. We start by reformulating web page analysis as finding expressions in extension of XPath. Then we propose to automatically discover these XPath expressions with InferXPath meta-language. Our goal is to automate laborious process of conversion of manually created web pages that serve as software documentations, wikis, and reference documents, and speed up their conversion into tabular data that can be directly fed into data pipeline.