Multimedia Knowledge and Social Media Analytics Laboratory

MOTOROLA funded projects

MKLab has collaborated with Motorola Research in a series of projects:

Recommender System Development via Automatic Semantic Induction

The project's scope is to develop a distributed semantic framework for personalised recommendation using advanced textual analysis and user profiling techniques,  knowledge-based reasoning and collaborative filtering to enable targeted recommendation of textual content, for any domain for which there is textual data even when little or no expert knowledge is available. A patent ("Ontology Construction") was filed after the completion of the project.

Motorola US Inc, (2009-2010)

Personalized Recommendation of Multimedia Content and Advertising by Means of Distributed Knowledge-based Reasoning

The aim of the project was to develop semantic textual analysis, knowledge-based reasoning and content filtering technologies to enable targeted advertising based on analysis of textual descriptions associated to advertisements, usage tracking, and implicit user profiling by means of machine learning techniques. A patent ("Method and System for Recommendation of Content Items") was filed, a paper was presented at the 2009 ACM RecSys conference and a book chapter was written after the completion of the project.

Motorola UK Ltd, (2007 - 2008)

Contextual Search based on Semantic Text Analysis

The project aimed at the development of a framework for the automatic formulation of search queries based on the current context (e.g. web page being read) of the user. The framework comprises a series of modules for semantic analysis, query formulation, query validation and query expansion. A patent ("Selection of Associated Content for Content Items") was filed and the following paper was presented at the MUM 2008 conference in Umea, Sweden.

Motorola UK Ltd, (2007)

Distributed Knowledge Management for Personalized Content Delivery

The main objective was to build a distributed reasoning architecture, aiming at delivering personalized news content to mobile users. The personalization process builds upon a text analysis framework for news content and the development of algorithms for adaptive user profiles. A patent ("A content recommendation system") was filed and a journal article was written after completion of the project.

Motorola UK Ltd, (2005 - 2006) 

 


 

The collaboration with Motorola has yielded a number of publications, namely 5 patents, 2 conference papers, a journal article and a book chapter:

Patents:

Conference publications:

Journal publications and book chapters:

D. Tsatsou, S. Papadopoulos, I. Kompatsiaris, P.C. Davis, "Distributed Technologies for Personalized Advertisement Delivery", in Online Multimedia Advertising: Techniques and Technologies, Xian-Sheng Hua, Tao Mei, Alan Hanjalic (editors), IGI Global, 2011.

Maria Papadogiorgaki, Vasileios Papastathis, Evangelia Nidelkou, Simon Waddington, Ben Bratu, Myriam Ribiere and Ioannis Kompatsiaris, "Two-level Automatic Adaptation of a Distributed User Profile for Personalized News Content Delivery", International Journal of Digital Multimedia Broadcasting, Special Issue on Personalization of Mobile Multimedia Broadcasting, vol. 2008, Article ID 863613, 21 pages, 2008. doi:10.1155/2008/863613.