Web2Text: Deep Structured Boilerplate Removal

TL;DR

This paper introduces a deep structured model that effectively removes boilerplate content from web pages, significantly improving main content extraction for NLP and IR tasks.

Contribution

It presents a novel sequence labeling approach using a hidden Markov model with CNN-derived features for boilerplate removal, achieving state-of-the-art results.

Findings

Sets new state-of-the-art on CleanEval benchmark

Improves retrieval performance on ClueWeb12

Effective classification of web page content blocks

Abstract

Web pages are a valuable source of information for many natural language processing and information retrieval tasks. Extracting the main content from those documents is essential for the performance of derived applications. To address this issue, we introduce a novel model that performs sequence labeling to collectively classify all text blocks in an HTML page as either boilerplate or main content. Our method uses a hidden Markov model on top of potentials derived from DOM tree features using convolutional neural networks. The proposed method sets a new state-of-the-art performance for boilerplate removal on the CleanEval benchmark. As a component of information retrieval pipelines, it improves retrieval performance on the ClueWeb12 collection.