Section: Other Grants and Activities

Regional Initiatives

Due to the bi-localization of the team, we are involved with two regions: PACA and Ile-de-France.

Color CusCOV (Inria, UNSA, CASA)

Participants : Julie Marlier, Bernard Senach, Brigitte Trousse.

A previous successfull collaboration during the MobiVIP project (See Delivrable 5.3. [102] ) with Agglomeration Community of Antibes Sophia Antipolis (CASA) gave opportunity for further work related to the usage analysis of a Web site dedicated to car-sharing;

CUsCov stands for "Confiance USages et COVoiturage". This project (2008-2009) starts with a Color Action aimed at analysing the role of trust in car-pooling.

Our partners are: LAMIC Laboratory (B. Conein) of University of Nice Sophia Antipolis (UNSA) and the transportation department of a local authority (Agglomeration Community of Antibes Sophia Antipolis) with the participation of M. Relieu (Paris Tech). The goal of the study is to better understand the barrier on car pooling, to judge how ICT can encourage this mobility solution and to find out which "social functions" could encourage to adopt car pooling. Considering that car pooling is a mixture of connecting technologies and of sociology (motivation and reluctance), ICT could contribute to ease the spreading of this mobility solution. One could expect that providing tools for social networking could support crew constitution and community emergence. More precisely, the project aims at understanding the role of trust in "crew" composition and to see if "social functionalities" such as detailed personal information or photography would help to set up a social network and contribute to car pooling spreading.

An internship Julie Marlier located at Inria Sophia Antipolis conducted a survey with users of car pooling combined with a log analysis and a web site analysis. She collected data through interviews and questionnaires and her results draw interesting conclusions, notably that :

• in a professional context, trust is not a main issue

• there are 3 main profiles : ecologist / pragmatic / social, each one with its own motivations

• most of the "social functions" proposed to the participants are rejected because "a car pooling site is not a meeting site" or because they "denatures the spirit of car-pooling"

• using car pooling as a transportation solution is not as easy as one could imagine : after a first contact , realizing the travels is not guarantee due to different timetable, lack of flexibility, specific constraints (scholars) . There are in fact many reasons for which car pooling becomes rapidly not feasible

• a lot has still to be done in car pooling web site to improve mapping between supply and demand and to support effective use of car pooling [69]

Color MUTAN (Inria, LGI2P Nimes, LIRMM)

Participants : Florent Masseglia, Gaurav Panthari, Goverdhan Singh, Dipankar Das, Nischal Verma.

MUTAN (2008, 3 internships) for “MUTualisation des ANomalies pour la détection d'intrusions” is a project on intrusion detection. The main goal of MUTAN is to lower the number of false alarms in intrusion detection based on anomalies. Actually, an anomaly is not always due to malicious behaviour and considering anomalies for intrusion detection is impossible to use because of the high rate of false positives. The main idea of MUTAN is that successful attacks will be repeated from one site to another and will be considered as anomalies. Therefore, if the same anomaly is detected on two (or more) different sites, then it is very likely to be a true intrusion. In 2008, partners of experimentation MUTAN have written 3 research reports [66] , [64] , [71] .

3 papers from this Color have been accepted in conferences in 2009 (EGC, PAKDD).

FOCUS Platform (PACA CPER TELIUS)

In 2007, in a framework agreement between french government and PACA territory (CPER Telius), AxIS proposed the creation of an experimentation platform FOCUS (Finding Out Collective Usage). This proposal was accepted and in 2008 AxIS received funding for buying equipments useful for future experiments in the Region. The FOCUS platform is dedicated to researchers interested in usage analysis, could be web usage, co-conception, user centred pilots in real life (scale 1) or experiments in laboratories contexts. Support offered to computer science community and industrial concerned by usage assessment, prototype validation and benchmarking will help them to improve practice in analyzing usage data.

The FOCUS platform relies on knowledge bases, data sets, data acquisition devices, software toolbox and specific methods for mining usage data. Main components of FOCUS platform are :

Data knowledge level is composed of :

• Data sets and benchmarks. The benchmarks available with the platform are well controlled data set that will be used by researchers to validate their analysis techniques,

• Experience data base : previous experimental results and conclusions in analyzing collective usage, past analysis cases for supporting new ones,

• Knowledge base: EAR methods, best practice, assessment criteria, recommendations, collective activities models.

Tool level is composed of devices and software resources

• Data acquisition tools : audio and video devices, advances tape recorder to synchronize heterogeneous data, eye-trackers, sounds to catch web usage,

• KDD toolbox : clustering and data stream mining algorithms, XML document checking,

• Experience reuse and IR support : object oriented framework for case based reasoning.

Multi-disciplinary approach: Stakeholders in FOCUS platform are coming from different disciplines: computer and social sciences (cognitive ergonomics, economy, sociology) which provide a very interesting confrontation of points of view and enriches usage analysis, for instance through multidisciplinary assessment criteria.

PACA Clusters or “Pôles de compétitivité”

• Cluster SCS - Solutions Communicantes Sécurisées: AxIS (B. Trousse and B. Senach) was involved in the preparation of the project Clic&Go. The goal of this project is to provide tools for proximity e-commerce. New technologies are used not to sell products on-line but to support new relationships between customers and shopkeepers. A better information will be given to the clients (availability of products with well-defined attributes - size, colors, etc.) and merchants will be able to develop loyalty through technologies. Usage mining will be used to improve user profiling and push suggestions according to similarity of buyers and profils. In Clic&Go project, academic partners are LSIS (UMR CNRS 6168), INRIA (AxIS) and industrial partners are STID, COMLINKS and AGEVIA. The project labelled by “Pôle de competitivité SCS” in June last year did not received a grant from the region. After few meetings in 2008 the consortium suspended its activities.

• Pegase Cluster with the CEFH Platform (Centre d'Etudes des Facteurs Humains): Areva TA, leader of the CEFH project contacted us for a collaboration in the set up of an experimentation platform dedicated to human factors studies in risky domains (nuclear power plant, aeronautics industry, process control. After a first contact in April, B. Senach participated in the first meeting in Aix en Provence, a common proposal was submitted to the call "Plateforme d'innovation" (DGE and CDC) The CEFH platform project is supported by the PEGASE cluster.

Living Lab “ICT Usage Lab”

Participants : Bernard Senach, Brigitte Trousse.

AxIS is involved with PACA partners (Orange labs, CSTB, UNSA, ...) in the set up of the Living lab.”ICT Usage Lab” (Sophia-Antipolis) who was labelled during the first wave by the European Network of Living Labs (ENoLL) in 2006 and was the first French living lab. We also were involved in a submission (called INFRALLABS) with almost 15 other Living Labs to the P7 call “Research Infrastructure” [20] . A booklet and a flyer are under construction.

Other initiatives

PacaLabs Initiative: AxiS took part in the presentation of the Pacalab financing structure. We also organised a meeting with the FING  (FING:)to envision future collaborations.

Two projects were submitted to Pacalab:

• PERFECT (PERFECT: PERFormance ECologique pour Trajets automobiles)with Accenture labs, ADEME and CASA: the goal of this project is to provide an information system aiding car drivers to reduce their gasoline consumption and to adopt driving behaviors contributing to sustainable development. Partners of the project are Accenture, ADEME, and CASA

• Hotel-Ref-PACA : the goal of the project is to identify from an experimental approach what can be the optimal location of commercial links and how to avoid bouncing (one visit pages). Partner are : Perferencement company and a local territory.

AxIS (B. Trousse and B. Senach) are involved in the preparation of a new project (AccésCité) aiming at providing guidance for thematic urban travelling. The device will be used, in particular by disabled people and usage mining tools will allow to improve information content delivered to users.