Skip to Main content Skip to Navigation
Theses

Linking and Mining Event-centric Data in the Semantic Web

Abstract : The widespread growth of social media has shifted the way people explore and share information of interest. Part of this evolution concerns the way our social activity is structured around events and increasingly documented by user-generated content. With the advent of Web 2.0, the vast amount of information about events, illustrative media and social interactions are spread and locked into diverse sites that provide a limited event coverage. In this thesis, we study the problem of integrating event-centric data into a unified environment, and we propose new approaches that could enhance content personalization in event-based services. The thesis is structured around two main parts dealing with key challenges related to the complex nature of events which are multifaceted, ephemeral and social entities. The first part focuses on event enrichment by leveraging semantic web technologies to integrate heterogeneous sources such as event websites, media platforms and social networks. Our strategy is to discover in real-time meaningful connections between events, venues, media an people. The second part tackles the problem of information overload. It investigates new personalization approaches helping individuals to discover interesting events and like-minded users. Our study underlines the important role of ontology-based content representation and collaborative filtering techniques to enhance event recommendation. We also propose a new solution to detect meaningful communities in event-based social networks by analyzing both the users’ interests and the structural properties of the underlying networks. The approaches proposed in this thesis have provided new basis for building an enhanced web environment changing the way to explore and organize events.
Document type :
Theses
Complete list of metadata

Cited literature [139 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/tel-01368243
Contributor : Houda Khrouf <>
Submitted on : Monday, September 19, 2016 - 11:58:13 AM
Last modification on : Thursday, February 11, 2021 - 9:19:33 AM

Licence


Copyright

Identifiers

  • HAL Id : tel-01368243, version 1

Collections

Citation

Houda Khrouf. Linking and Mining Event-centric Data in the Semantic Web. Computer Science [cs]. LTCI - Laboratoire Traitement et Communication de l'Information [Paris], 2014. English. ⟨tel-01368243⟩

Share

Metrics

Record views

33

Files downloads

49