Representation models for text classification: A comparative analysis over three Web document types

Giannakopoulos, G; Mavridi, P; Paliouras, G; Papadakis, G; Tserpes, K

dc.contributor.author	Giannakopoulos, G	en
dc.contributor.author	Mavridi, P	en
dc.contributor.author	Paliouras, G	en
dc.contributor.author	Papadakis, G	en
dc.contributor.author	Tserpes, K	en
dc.date.accessioned	2014-03-01T02:54:00Z
dc.date.available	2014-03-01T02:54:00Z
dc.date.issued	2012	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/36530
dc.subject	N-gram graphs	en
dc.subject	Text classification	en
dc.subject	Web document types	en
dc.subject.other	Bag of words	en
dc.subject.other	Comparative analysis	en
dc.subject.other	Contextual information	en
dc.subject.other	Experimental studies	en
dc.subject.other	Feature space	en
dc.subject.other	Graph similarity	en
dc.subject.other	N-gram graphs	en
dc.subject.other	News articles	en
dc.subject.other	Real world data	en
dc.subject.other	Representation model	en
dc.subject.other	Sentiment analysis	en
dc.subject.other	Social media	en
dc.subject.other	Spam filtering	en
dc.subject.other	Text classification	en
dc.subject.other	Textual content	en
dc.subject.other	Topic Classification	en
dc.subject.other	User-generated content	en
dc.subject.other	Web document	en
dc.subject.other	Graphic methods	en
dc.subject.other	Information retrieval systems	en
dc.subject.other	Semantic Web	en
dc.subject.other	Semantics	en
dc.subject.other	Virtual reality	en
dc.subject.other	World Wide Web	en
dc.title	Representation models for text classification: A comparative analysis over three Web document types	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1145/2254129.2254148	en
heal.identifier.secondary	http://dx.doi.org/10.1145/2254129.2254148	en
heal.identifier.secondary	13	en
heal.publicationDate	2012	en
heal.abstract	Text classification constitutes a popular task in Web research with various applications that range from spam filtering to sentiment analysis. To address it, patterns of co-occurring words or characters are typically extracted from the textual content of Web documents. However, not all documents are of the same quality; for example, the curated content of news articles usually entails lower levels of noise than the user-generated content of the blog posts and the other Social Media. In this paper, we provide some insight and a preliminary study on a tripartite categorization of Web documents, based on inherent document characteristics. We claim and support that each category calls for different classification settings with respect to the representation model. We verify this claim experimentally, by showing that topic classification on these different document types offers very different results per type. In addition, we consider a novel approach that improves the performance of topic classification across all types of Web documents: namely the n-gram graphs. This model goes beyond the established bag-of-words one, representing each document as a graph. Individual graphs can be combined into a class graph and graph similarities are then employed to position and classify documents into the vector space. Accuracy is increased due to the contextual information that is encapsulated in the edges of the n-gram graphs; efficiency, on the other hand, is boosted by reducing the feature space to a limited set of dimensions that depend on the number of classes, rather than the size of the vocabulary. Our experimental study over three large-scale, real-world data sets validates the higher performance of n-gram graphs in all three domains of Web documents. Copyright 2012 ACM.	en
heal.journalName	ACM International Conference Proceeding Series	en
dc.identifier.doi	10.1145/2254129.2254148	en