%0 Conference Paper %1 10.1007/978-3-319-76941-7_13 %A Vogels, Thijs %A Ganea, Octavian-Eugen %A Eickhoff, Carsten %B Advances in Information Retrieval %C Cham %D 2018 %E Pasi, Gabriella %E Piwowarski, Benjamin %E Azzopardi, Leif %E Hanbury, Allan %I Springer International Publishing %K boilerplate tool web2text %P 167--179 %T Web2Text: Deep Structured Boilerplate Removal %X 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. %@ 978-3-319-76941-7

@inproceedings{10.1007/978-3-319-76941-7_13, 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.}, added-at = {2020-12-11T16:22:05.000+0100}, address = {Cham}, author = {Vogels, Thijs and Ganea, Octavian-Eugen and Eickhoff, Carsten}, biburl = {https://www.bibsonomy.org/bibtex/244b2df10925e30a38acff298b4e5291c/parismic}, booktitle = {Advances in Information Retrieval}, description = {Web2Text: Deep Structured Boilerplate Removal | SpringerLink}, editor = {Pasi, Gabriella and Piwowarski, Benjamin and Azzopardi, Leif and Hanbury, Allan}, interhash = {4e013a030582b690116556a7156ad8bf}, intrahash = {44b2df10925e30a38acff298b4e5291c}, isbn = {978-3-319-76941-7}, keywords = {boilerplate tool web2text}, pages = {167--179}, publisher = {Springer International Publishing}, timestamp = {2020-12-11T16:22:05.000+0100}, title = {Web2Text: Deep Structured Boilerplate Removal}, year = 2018 }