Belgium: PH.D. student job offers in IR at K.U. Leuven (deadline Aug 19) (KDnuggets News 07:15, item 25, Academic/Research Positions)

KDnuggets : News : 2007 : n15 : item25

Academic/Research Positions

From: Sien Moens
Date: 02 Aug 2007
Subject: Belgium: PH.D. student job offers in IR at K.U. Leuven (deadline Aug 19)

Job Offers in Multimedia Summarization, Alignment of Multilingual and Multimedia Information, Spam E-mail Filtering, Question Answering Information Retrieval

Multimedia Summarization

We offer a position for a researcher working towards a Ph.D. in computer science on the topic of multimedia summarization. The research is partly performed in the frame of the EU-FP6 CLASS (Cognitive-Level Annotation Using Latent Statistical Structure) project (http://class.inrialpes.fr/). The goal is to summarize video broadcasted by different news channels (e.g., BBC, CNN, ABC) while jointly considering content from video images and accompanying text. The Ph.D. candidate will research models for extraction of salient content by exploiting multimodality and use these models for generating textual descriptions for the summarized video. The candidate has a strong interest in text summarization, natural language processing, text generation and probabilistic topic models, and has an affinity with the domain of image recognition. In this research we collaborate with the Department of Electrical Engineering (Visics) of the K.U.Leuven, INRIA, Grenoble (France), University of Oxford (UK), University of Helsinki (Finland) and Max-Planck Institute for Biological Cybernetics (Germany).

Alignment of Multilingual and Multimedia Information

We offer a position for a researcher working towards a Ph.D. in computer science on the topic of alignment of multilingual and multimedia information. The research is performed in the frame of the AMASS++ (Advanced Multimedia Alignment and Structured Summarization) project (http://www.cs.kuleuven.be/~liir/projects/amass/index.html), where we build crossmedia and cross-lingual summaries of content and present these in a user-friendly way. The summaries rely on automatic content recognition of the visual and textual medium that is reinforced by their mutual and integrated processing. Research in natural language processing that focuses on the alignment of equivalent content across texts (possibly written in different languages) is a source of inspiration for innovative models for cross-media alignment. As a test case we process the news archives of Flemish, UK and US broadcasters. The candidate has a strong interest in statistical machine translation and natural language processing in general, and has an affinity with the domain of image recognition. In this research we collaborate with the Department of Electrical Engineering (Visics) and the Centre for Computational Linguistics of the K.U.Leuven, and the Expertise Centre for Digital Media of the University of Hasselt (Belgium).

Spam E-mail Filtering

We offer a position for a researcher working towards a Ph.D. in computer science on the topic of spam e-mail filtering. The research is partly performed in the frame of the EU-FP6 AntiPhish (Anticipatory Learning for Reliable Phishing Prevention) project (http://www.antiphishresearch.org/partners.html), where we build filters for phishing mails. The focus is on natural language processing of the texts of the e-mails and the summarization of the mails in terms of the general and specific topics treated, and on developing models for predicting future phishing topics based on phishing contexts. The candidate has a strong interest in machine learning, natural language processing and probabilistic topic models. The research is performed in collaboration with Fraunhofer Institute for Intelligent Analysis and Information Systems (IAIS), Symantec (GB and IRL), Tiscali Group (Italy), and Nortel Networks S.A. (France).

Question Answering Information Retrieval

We offer a position for a researcher working towards a Ph.D. in computer science on the topic of question answering information retrieval. The research is performed in the frame of the European ITEA 2 LINDO (Large scale distributed INDexation of multimedia Objects) project (Web site of the project is under development). The research focuses on the querying of multimedia content by means of questions posed in natural language, and on ranking algorithms that compute the relevancy of a multimedia object for the query where the multimedia objects are represented as topic maps (ISO standard). The designed and developed technology will be applied for the retrieval of information from technical documentation for astronauts, surveillance videos and satellite images. The candidate has a strong interest in question answering technology, graph based algorithms and natural language processing. The work is done in collaboration with the companies Space Applications (Belgium) and Linguit (UK).

The successful candidate has a strong background in informatics, computer science or artificial intelligence.

Contact:
We offer a job in a growing and dynamic team with international reputation in collaboration with key members of academia, industry and governments, and give the candidate the opportunity to present the research results at multiple international conferences. The gained expertise will give him or her also a solid basis for a future career with considerable academic and economic potential in the information searching and mining market. We offer Ph.D. students a very competitive wage. Our research group (www.law.kuleuven.be/icri/liir.php ) moves to the Department of Computer Science of the Katholieke Universiteit Leuven before October 1, 2007. All jobs will start October 1, 2007 or earlier.

Send your questions, CV and motivation letter to Marie-Francine Moens (mariefrance.moens@law.kuleuven.be) the latest August 19, 2007.

KDnuggets : News : 2007 : n15 : item25

PREVIOUS | NEXT