Harnessing linked knowledge sources for topic classification in social media

Cano, Amparo Elizabeth; Varga, Andrea; Rowe, Matthew; Ciravegna, Fabio; He, Yulan

doi:10.1145/2481492.2481497

Cited by 18 publications

(36 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, in contrast to our previous work ( [13]), in this paper our main focus is to understand the differences between the different semantic concept graphs, and present a comparative evaluation of these graphs at different stages of our three-stage approach. The main stages of our approach can be summarised as follows: i) context modelling; ii) topic classification and iii) topic similarity analysis.…”

Section: Contributionsmentioning

confidence: 99%

“…In light of the classification we used in Cano et al [13], we distinguish between two main groups: approaches exploiting local features, and approaches exploiting the link structure of KSs.…”

Section: Semantic Linking Using Social Knowledge Sourcesmentioning

confidence: 99%

“…There is only little work which exploits multiple, linked KSs ( [13,14]). In Varga et al [14], we presented an approach which makes use only of the data within KSs, ignoring the semantic information about entities present in KSs.…”

Section: Semantic Linking Using Social Knowledge Sourcesmentioning

confidence: 99%

“…bag-of-words (BoW)) representation of text by incorporating additional contextual information about Microposts in the form of semantic (bag-of-entities (BoE)) features extracted from the content of Microposts only. Unlike these approaches, recently in [13] we proposed a TC framework which generates contextual information from graph structures surrounding concepts in multiple complementary linked KSs. Among the several useful graph structures defined in KSs ( [8]), such as the resource meta graph providing course grained classification of concepts by their type, or the category meta graph which groups similar concepts together by their topic, our original framework exploited the resource metagraph for context generation.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Linked knowledge sources for topic classification of microposts: A semantic graph-based approach

Varga

Basave

Rowe

et al. 2014

Journal of Web Semantics

Self Cite

View full text Add to dashboard Cite

Short text messages a.k.a Microposts (e.g. Tweets) have proven to be an effective channel for revealing information about trends and events, ranging from those related to Disaster (e.g. hurricane Sandy) to those related to Violence (e.g. Egyptian revolution). Being informed about such events as they occur could be extremely important to authorities and emergency professionals by allowing such parties to immediately respond.In this work we study the problem of topic classification (TC) of Microposts, which aims to automatically classify short messages based on the subject(s) discussed in them. The accurate TC of Microposts however is a challenging task since the limited number of tokens in a post often implies a lack of sufficient contextual information.In order to provide contextual information to Microposts, we present and evaluate several graph structures surrounding concepts present in linked knowledge sources (KSs). Traditional TC techniques enrich the content of Microposts with features extracted only from the Microposts content. In contrast our approach relies on the generation of different weighted semantic meta-graphs extracted from linked KSs. We introduce a new semantic graph, called category meta-graph. This novel meta-graph provides a more fine grained categorisation of concepts providing a set of novel semantic features. Our findings show that such category meta-graph features effectively improve the performance of a topic classifier of Microposts.Furthermore our goal is also to understand which semantic feature contributes to the performance of a topic classifier. For this reason we propose an approach for automatic estimation of accuracy loss of a topic classifier on new, unseen Microposts. We introduce and evaluate novel topic similarity measures, which capture the similarity between the KS documents and Microposts at a conceptual level, considering the enriched representation of these documents.Extensive evaluation in the context of Emergency Response (ER) and Violence Detection (VD) revealed that our approach outperforms previous approaches using single KS without linked data and Twitter data only up to 31.4% in terms of F1 measure. Our main findings indicate that the new category graph contains useful information for TC and achieves comparable results to previously used semantic graphs. Furthermore our results also indicate that the accuracy of a topic classifier can be accurately predicted using the enhanced text representation, outperforming previous approaches considering content-based similarity measures.

show abstract

Section: Contributionsmentioning

confidence: 99%

Section: Semantic Linking Using Social Knowledge Sourcesmentioning

confidence: 99%

Section: Semantic Linking Using Social Knowledge Sourcesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Linked knowledge sources for topic classification of microposts: A semantic graph-based approach

Varga

Basave

Rowe

et al. 2014

Journal of Web Semantics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The Semantic Web approach is about adding formal structures and semantics (metadata and knowledge) to Web content for easy access, management, discovery and integration, to make the resources machine-understandable. Some proposals, as Cano, Varga, Rowe, He and Ciravegna (2013) or Husby and Barbosa (2012) are based on the use of repositories of linked open data to determine the topics that describe social content as micropost or blogs. These and other studies have found similar findings, "DBpedia resources are a good starting point to define keyword meanings due to the fact that a huge part of the knowledge base is related to classes in the DBpedia Ontology" (Cano et al, 2013).…”

Section: Related Workmentioning

confidence: 99%

Open Praxis, volume 6 issue 2

Gil‐Jaurena¹

2014

Open Praxis

View full text Add to dashboard Cite

Towards a Better Contextualization of Web Contents via Entity-Level Analytics

Kumar

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

With the abundance of data and wide access to the internet, a user can be overwhelmed with information. For an average Web user, it is very difficult to identify which information is relevant or irrelevant. Hence, in the era of continuously enhancing Web, organization and interpretation of Web contents are very important in order to easily access the relevant information. Many recent advancements in the area of Web content management such as classification of Web contents, information diffusion, credibility of information, etc. have been explored based on text and semantic of the document. In this paper, we propose a purely semantic contextualization of Web contents. We hypothesize that named entities and their types present in a Web document convey substantial semantic information. By extraction of this information, we aim to study the reasoning and explanation behind the Web contents or patterns. Furthermore, we also plan to exploit LOD (Linked Open Data) to get a deeper insight of Web contents.

show abstract

Harnessing linked knowledge sources for topic classification in social media

Cited by 18 publications

References 11 publications

Linked knowledge sources for topic classification of microposts: A semantic graph-based approach

Linked knowledge sources for topic classification of microposts: A semantic graph-based approach

Open Praxis, volume 6 issue 2

Towards a Better Contextualization of Web Contents via Entity-Level Analytics

Contact Info

Product

Resources

About