- Savoir modéliser un comportement, une architecture, une structure à l’aide des graphes.
- maîtrise des méthodes de raffinement de modèles et de modélisation de contraintes pour mieux résoudre un problème.
Techniques :
Maîtrise des outils de graphes (structures et paramètres) et des aspects algorithmiques avancés (distribués, dynamiques, auto-stabilisants, «online »…) pour la modélisation et la résolution de problèmes.
Name of the course: Data Mining Number of ECTS credits:
Contact :
Exam Prepared lecture on a couple of influential data mining papers.
Course Content Data Mining has been identified as one of the ten emergent technologies of the 21st century (MIT Technology Review, 2001). This discipline aims at discovering knowledge from large amounts of data and its development has emerged at the intersection of various disciplines related to data processing, e.g., machine learning, database management, visualization, statistics. In a first part, we will provide an overview of the quite active research field of data mining and knowledge discovery in databases. Classical techniques (clustering and supervised classification, pattern discovery) will be considered. Examples of real-life data mining applications will concern, among others, basket data analysis, WWW usage data analysis, and knowledge discovery in living sciences (e.g., molecular biology).
A second part will be dedicated to constraint-based data mining and the emerging framework of inductive querying. After an introduction to this appealing formal framework, we will discuss some recent research topics related to the condensed representations of frequent patterns and constraint-based mining of sequential patterns. C1 KDD: motivations and terminology (Boulicaut)
C2 Data (Rigotti)
C3 Data exploration (Rigotti)
C4 Clustering (Rigotti)
C5 Classification (Rigotti)
C6 Association analysis (Boulicaut)
C7 Towards a theory of data mining (Boulicaut)
C8 Condensed representations for frequent patterns (Boulicaut)
C10 A research agenda(Boulicaut) The course is based on the excellent book by Pang-Ning Tan, Michael Steinbach and Vipin Kumar “Introduction to data mining” published in 2006 by Addison-Wesley (slides have been provided by the authors).
It will be possible to apply the techniques on benchmark data by using the software platform Weka (free software).
Targeted Skills The popular techniques for data mining (e.g., K-Means and hierarchical clustering, decision tree building, association rule mining) are understood. Some recent conceptual issues or data mining principles related to inductive querying and constraint-based mining are understood as well: it provides a conceptual framework for analysing the current research directions in data mining.
Name of the course: Pervasive information systems Number of ECTS credits:
Contact :
Name and Given names : Frederique LAFOREST
Phone : 04 72 43 89 83
email : Frederique.laforest@liris.cnrs.fr
Other professor(s) if any :
Exam
Exam during 2 hours
Course Content I Introduction
II Distributed and pervasive information systems architecture
III.3.4 - adaptation of data, user interfaces and services
IV Exemples of pervasive projects
V Conclusion
Targeted Skills Methodological : study and critics of the littérature, bibliography
Technical : architecture of distributed, mobile and pervasive systems, mobility management techniques (disconnection, caching…), of context (capture, interpretation, modelisation), of adaptation (user interfaces, data and services)
Name of the course: Visual Information Systems Number of ECTS credits: 3
Contact :
Name and Given names : Robert Laurini
Phone : +33 (0)4 72 43 81 72
email : +33 (0)4 72 43 87 13
Exam Written exam.
Course Content This course is devoted to visual information systems with both meanings, i.e. Systems of Visual Information, and Visual Systems of Information.
So, the objective of this course is to give the students the fundamentals to understand and to design information systems using images, raster or vector, not only for graphic user interfaces, but also on image contents. Especially, will be studied geographic information systems, images information systems, portals, etc.
The contents are as follows:
1 – Visions of Space
8 – Visual Portals to multimedia Information Systems
9 – Contents Visualization
10 – Image Extension of ORACLE
Prerequisites : database design, information systems design, computational geometry, image processing
Nom de l’UE : Physical layer modelling for future wireless networks Nombre de crédits :
Contact :
Nom & Prénom(s) : Gorce Jean-Marie
Tél. : 04 72 43 60 68
email : jean-marie.gorce@insa-lyon.fr
Autre(s) intervenant(s) :
Mischa Dohler, France Télécoms R&D, Grenoble
Contrôle des connaissances : An oral presentation and a written work on published papers
Programme – contenu détaillé de l’UE Introduction: After a fast development during the last decade, wireless networks are everywhere. While cellular networks allow a large-scale mobility, local networks such as WLAN offer a friendly wireless link in local areas. During the near past, new advances focused on improving each technology exploiting a reserved frequency bandwidth.
The future opens new trends. Future technologies will have to share the same radio resource and to fit the quality of the radio environment: the cognitive or opportunistic radio is ongoing. This new approach will offer a wide set of new services: the future terminal will be multi-mode, reconfigurable, and less energy consuming.
To achieve this mutation, the future terminal will have to comply with the complexity of the real wireless medium. It is true that radio offers mobility and friendly wireless access, but in turns exhibits strong problems: packet error rate, fading, interference.
To manage the wireless medium, the standard network model (OSI model) has been enhanced by introducing the MAC layer which allows the data link layer to manage the physical layer. The MAC layer introduces rules to share efficiently the medium between terminals.
Lot of routing and MAC protocols are evaluated on the basis of a very simple physical layer model (circular, threshold). Some of them may fail when used in a realistic environment.
To improve these protocols, a better knowledge about the physical layer is needed. New ideas will probably come from cross-layer studies and cross-layer protocols. MAC-PHY protocols are those taking into account carefully the reality of the physical medium.
To understand these trends, the students working on protocols for wireless networks should first acquire basics of the physical layer. Secondly, they should be aware of mathematical tools for modelling this physical layer. This course aims to help students in computer sciences to improve their knowledge about both.
Scheduling :
The mistaken axioms of wireless network (2h, JMG)
What are the common assumptions about wireless networks and why they are false?
Propagation modelling (2h, MDO)
The common models are described: path-loss models and shadowing and fading.
Modulation, BER and radio link quality (2h, JMG)
Starting from the standard circular threshold model, real assumptions are studied: a BER (bit error rate) is introduced. Channel coding effect, …
Interference (2h, JMG)
What are the main laws of interference : equivalent noise, rejection capability,
How interference can be introduced in wireless networks? More specifically in ad hoc or sensor networks.
Resource sharing (4h, MDO)
Because lot of recent works are devoted to multi-channel techniques, it is important to detail how a frequency band can be shared: TDMA/FDMA/CDMA/OFDMA/…
Lot of works assume a perfect orthogonality between sub-channels. This is not true. How the multi-channels interfere will be modelled.
A practical study based on a wLAN deployment (2h, JMG)
A complete development on how these assumptions can be introduced in a wLAN modelling
A practical study for ad hoc/sensor networks (2h, MDO)
Introducing some of these realistic assumptions in ad hoc network modelling is discussed
Scientific papers presentation (4h, JMG/MDO) :
Students will present some articles about the topic of the course. 3 presentations per hour up to 12.
Compétences acquises Méthodologiques :
Realistic modelling of the physical layer of radio networks
