Retrieving Informative Content from Web Pages with Conditional Learning of Support Vector Machines and Semantic Analysis

Specification and discovery of web patterns: a graph grammar approach

“…The Wrapper-based approaches, such as [17,19,38], induce a set of rules to define the knowledge of data extraction. Statistical models (such as Support Vector Machines [30] or Hidden Markov Models [21,47]) discover repetitive structures based on statistical analysis. Distinct from wrappers or statistical models, our approach applies the state-of-the-art grammar technology (i.e.…”