Supervised and Unsupervised Methods for Robust Separation of Section Titles and Prose Text in Web Documents

Abstract

The text in many web documents is organized into a hierarchy of section titles and corresponding prose content, a structure which provides potentially exploitable information on discourse structure and topicality. However, this organization is generally discarded during text collection, and collecting it is not straightforward: the same visual organization can be implemented in a myriad of different ways in the underlying HTML. To remedy this, we present a flexible system for automatically extracting the hierarchical section titles and prose organization of web documents irrespective of differences in HTML representation. This system uses features from syntax, semantics, discourse and markup to build two models which classify HTML text into section titles and prose text. When tested on three different domains of web text, our domainindependent system achieves an overall precision of 0.82 and a recall of 0.98. The domaindependent variation produces very high precision (0.99) at the expense of recall (0.75). These results exhibit a robust level of accuracy suitable for enhancing question answering, information extraction, and summarization. 1

Utilizing Sentence Embedding for Dangerous Permissions Detection in Android Apps' Privacy Policies

“…Sentenceembeddingmodelsarepromisingtechniquesthatareusedtocapturesentencessemantics andtheirrelations.Therearedifferentapplicationsthatrelyonencodingsemanticmeaningofprivacy policiessentences,suchasapplicationsinterestedincheckingAndroidapps'behaviorsagainstwhat isstatedintheirprivacypolicies. Infact,manyprivacyrelatedapplicationssuchasLiu,Fella,& Liao(2016), Gopinath,Wilson,&Sadeh(2018), Sun(2018),andHarkousetal. (2018usewordor sentenceembeddingmodelsaspartoftheautomaticanalysisofprivacypolicies.However,itisnot entirelycleartowhatextentsentenceembeddingsareeffectiveincapturingthesemanticsofprivacy policiessentences.Therefore,toensurethesuccessfulnessofsuchapplications,itiscrucialtoreport theadvantagesanddisadvantagesofusingsentenceemebddingsandsuggestimprovementsifneeded.…”

A Scalable Framework for Table of Contents Extraction from Complex ESG Annual Reports

“…Tuarob et al (2015) designed some features and use Random Forest (Breiman, 2001) and Support Vector Machine (Bishop and Nasrabadi, 2006) to predict section headings. Mysore Gopinath et al (2018) propose a system for section titles separation. MTD represents a more recent approach, fusing text, visual, and layout information to detect section headings from scientific papers in the Hi-erDoc dataset.…”

Breaking Down Walls of Text: How Can NLP Benefit Consumer Privacy?

“…fectively also requires several additional capabilities such as reasoning over vagueness and ambiguity, understanding elements such as lists (including when they are intended to be exhaustive and when they are not (Bhatia et al, 2016)), effectively incorporating 'co-text'-aspects of web document structure such as document headers that are meaningful semantically to the content of privacy policies (Mysore Gopinath et al, 2018) and incorporating domain knowledge (for example, understanding whether information is sensitive requires background knowledge in the form of applicable regulation). Privacy policies also differ from several closely related domains, such as legal texts which are largely meant to be processed by domain experts.…”

“…There have also been efforts to analyze vague statements in privacy policies Lebanoff and Liu, 2018), and explore how benchmarks in this domain can be constructed through crowdsourcing Wilson et al, 2016c;Audich et al, 2018). Lastly, there has been research focused on identifying header information in privacy policies (Mysore Gopinath et al, 2018) and generating them (Gopinath et al, 2020 (Harkous et al, 2018) 120…”