The Business Intelligence (BI) Practice at SAP Research is looking for a strong candidate for a PhD position in the area of graph visualization/filtering/clustering.
The BI Practice is part of the global technology research unit of SAP and works on next-generation data access and information delivery tools. Our mission is to bring breakthrough innovation and research, and contribute to the success of SAP’s in user-centric BI and analytics technologies. In order to contribute to the Information Visualization research agenda, the BI Practice proposes a PhD position in its Sophia-Antipolis offices (Nice, France).
PhD TOPIC DESCRIPTION:
Recent trends in information and knowledge modeling are increasingly oriented towards graph based models that capture the relational structure between diverse types of entities. Areas ranging from knowledge and model-driven engineering, information retrieval, semantic technologies to social networks and unstructured information analysis are just a few examples where graph-based models have been successfully applied. Although relational information models have held their promise in taking knowledge modeling a step further, in practice, there is a major difficulty to their efficient use: visual complexity. In particular for large, strongly connected graphs it is not only difficult to visualize all information at once, but also to decide how to zoom-into and explore the data without losing a general overview of the context.
Filtering and aggregating information is an important line of research to overcome the above cited drawbacks. They can be envisaged both as pre-processing steps and as real-time mechanisms in view of allowing interactive, user-oriented visual analysis of graphs. Current approaches to graph filtering essentially use random or purely structural measures on graphs. Such techniques are however inadequate to process graphs that model knowledge structures, since prone to distort the knowledge model by disregarding the information content. State-of-the-art graph aggregation techniques, in turn, face similar limitations: they can be purely content-based, i.e. clustering oriented, or purely structural. Recent proposals have started to focus on mixed approaches to improve user-experience on the exploration of document databases.
In brief, efficient information visualization is just as much about knowing what to show than knowing what not to show. A first important research point is to test whether existing approaches for graph filtering/aggregation are suited to the knowledge structures that arise in intelligent business applications – and whether their implementation allows for real-time interaction for visual analysis purposes. Second, it is worth exploring to what extent the success of these techniques is problem specific and how to develop more general and robust approaches. Finally, it is of interest to investigate entirely new approaches that build upon graphs as "models" for the filtering process. Such an approach, inspired by recent work in graph matching and graph querying, would make it possible to take into account both the structure and the content of the knowledge bases, as well as to incorporate context and user information to customize the visual analysis experience.
· Last year student of M.Sc. or Engineering degree in Computer Science, Applied Mathematics or similar fields. Students with a strong academic record are favored.
· Interest for applied research and development. The candidate should be comfortable with theoretical / abstract work as well.
· Knowledge in Data Mining, Operations Research, Graph Theory and related fields. Ideal candidate has knowledge of algorithms for graph analysis and visualization, both theoretical and practical.
· Experience with programming languages (Java)
· Fluency in English (working language), other EU languages considered an advantage (in particular French)
· Excellent oral and written communication skills
As the world's leading provider of business software, SAP delivers products and services that help accelerate business innovation for our customers. We believe that doing so will unleash growth and create significant new value – for our customers, SAP, and ultimately, entire industries and the economy at large. Today, more than 47,800 customers in more than 120 countries run SAP applications – from distinct solutions addressing the needs of small businesses and midsize companies to suite offerings for global organizations.
As the global technology research unit of SAP, SAP Research significantly contributes to SAP's product portfolio and extends SAP's leading position by identifying and shaping emerging IT trends through applied research and corporate venturing. In contrast to SAP's product groups, which work on new functions and releases, SAP researchers explore opportunities that haven't yet been developed into products. We track technology trends, evaluate the potential impact on SAP solutions and customers, and generate breakthrough technologies. The business model of SAP Research is based on co-innovation through collaborative research. Working with leading universities, partners, customers, and SAP product groups, SAP Research oversees the development of promising ideas and prototypes into market-ready software for maximum customer value.
SAP Research Sophia Antipolis is located in the setting of one of the most important scientific parks worldwide. A high concentration of IT and telecommunication industries within walking distance, proximity to partners and customers, as well as the educational establishments in the region provide SRC Sophia Antipolis with an ideal working environment. Based at SAP Labs France, SRC Sophia Antipolis focuses on the topics of security and trust (main focus) and information management (in development).
Please send your CV and relevant documents to Géraldine Bous ([log in to unmask]<mailto:[log in to unmask]>, tel. +33-(0)4-92286332). Make sure your e-mail has header: [PhD Application] Intelligent Filtering for Graph Visualization.
To unsubscribe, send an empty email to
mailto:[log in to unmask]
For further details of CHI lists see http://listserv.acm.org