Personalised AdveRtisements buIlt from web Sources
User generated content, e.g., available through social networking sites on the World Wide Web, offers a wealth of information. The aim of PARIS is to study adequate natural language, image and video understanding techniques to make use of this information. This is challenging as user generated content is often present in language that is not well-formed or in visual material that is not professionally captured. An important application of PARIS when evaluating the natural language, image and video understanding techniques is personalized advertising, where advertisements are searched and generated that fit the world of living of users. This personalization should remain in line with legal requirements concerning personal data protection and intellectual property rights.
PARIS aims here at the smooth integration of content, advertisements and users by means of novel information and communication technology. However, many other applications of the PARIS technologies are possible, for instance, aimed at mining societal trends and their impact.
LIIR is involved in the natural language processing research, the cross-media mining of user generated content and in retrieval models for advertisement retrieval.
PartnersLIIR coordinates the PARIS project. The other partners are: KU Leuven ESAT-Visics (Prof. Luc Van Gool, Prof. Tinne Tuytelaars), KU Leuven HCI (Prof. Erik Duval), the Computational Web Intelligence group of Ghent University (Prof. Martine De Cock), the Language and Translation Technology Team of HoGent (Prof. Veronique Hoste), the International and European Legal Studies Programme of the University of Antwerp (Prof. Patrick Van Eecke) and the Vlerick Leuven Gent Management School (Prof. Steve Muylle).
An edited book on mining of user generated content is in print.
We have obtained first results in personality recognition given textual user generated content.
|Period||From 2012-05-01 to 2016-04-30.|
|Financed by||IWT-SBO-Nr. 110067|
|Supervised by||Marie-Francine Moens|
|More information can be found on the project website http://www.parisproject.be/|
- DE SMET, Wim & MOENS, Marie-Francine Representations for Multi-Document Event Clustering. Data Mining and Knowledge Discovery, 26(3): 533-558. 2013
- KOLOMIYETS, Oleksandr & MOENS, Marie-Francine MotionML: Motion Markup Language: Shallow Approach for Annotating Motions in Text. In Proceedings of Corpus Linguistics. 2013
- FARNADI, Golnoosh, ZOGHBI, Susana, MOENS, Marie-Francine & DE COCK Martine Crawling a Ground Truth Dataset for Ad Personalization. In Abstracts of LTCI2012 (5th AUGent Workshop on Language Technology and Computational Intelligence), p.7. 2012
- FARNADI, Golnoosh, ZOGHBI, Susana, MOENS, Marie-Francine & DE COCK Martine Recognising Personality Traits using Facebook Status Updates. In Proceedings of WCPR13, Workshop on Computational Personality Recognition at ICWSM13 (7th International AAAI Conference on Weblogs and Social Media). 2013
Back to all projects