Sofiane LAGRAA's Homepage

PhD Post-Doc

About me
Current position
I am a Post-doc researcher at the Loria lab. as well as at Inria Nancy - Grand Est within Madynes team, under the supervision of Jérôme François. I work on behavior-based security analytics using data mining and graph theory.
FUI project: HUMA "Human into the loop of Massive security data analysis"
Previous positions
  • Post-doc researcher at the CITI laboratory as well as at INRIA Rhône-Alpes within PRIVATICS team, under the supervision of Marine Minier. I worked on representation and knowledge extraction from a huge volume of data(logs) using data mining for security data analysis. full time 2015-2016
    • FUI project: HUMA "Human into the loop of Massive security data analysis".
  • Teaching assistant (ATER in french) in computer science at IUT2 of Grenoble full time 2014-2015
  • Teaching assistant (ATER) in computer science at University Grenoble I of Grenoble partiel time 2013-2014
  • PhD in computer science at Grenoble INP under the supervision of Frédéric Pétrot and Alexandre Termier within TIMA and LIG laboratories, respectively. Oct. 2010 - June.2014
  • Teaching assistant (Monitorat in french) in computer science at University Grenoble I of Grenoble partiel time 2011-2013
  • Master in computer science at INSA of Lyon. Specialty: Artificial intelligence 2009-2010
  • Master thesis under the supervision of Hamida Seba, within Graphs, algOrithms and AppLications (GOAL) research group of the LIRIS lab.
Research interest
My main research interest is:
  • Pattern recognition and Knowledge discovery using Data Mining and Graph theory algorithms applied to:
    • Data analysis, traces (logs) analysis, profiling
  • (Large) Graph Matching and (Large) Graph Analytics
  • Embedded System: MPSoC, Parallel Application,
  • Security: attack detection: port-scan, DDoS, behavior analytics, root cause of attacks.
PhD Thesis
Manuscript thesis
The goal of my thesis is to define and design new profiling tools for the evaluation of performance and power for applications mapped on Multi-Processor System-on-Chip. As the number of processors and tasks increases in embedded applications, being able to automatically extract hot spot or spot dysfunctional behaviors from gigabytes of traces becomes both an interesting scientific challenge and a practical issue. The approaches based on Data Mining are the appropriate choice for digging into the large amount of data to process. For example, Frequent Pattern Mining allows to discover patterns having a complex structure (sequences, trees or graphs) occurring frequently in data. The thesis will study the applicability of the Frequent Pattern Mining techniques, and clustering for analysis execution traces extracted from simulated embedded parallel program executions on the SoC and propose solution to make it practical. It also targets to adapt these techniques in order to discover problematic access patterns such as memory contention or bad temporal locality and scalability bottlenecks.
Publications
My profile on DBLP
Conferences
Year Paper
2017 Sofiane Lagraa, Jérôme François, Abdelkader Lahmadi, and Marine Minier. Mining Behavior Heterogeneous Network Event Logs for Root Cause Analysis. [To be Submitted]
2017 Sofiane Lagraa, Jérôme François. Knowledge Discovery of Port Scans from Darknet. IFIP/IEEE International Workshop on Analytics for Network and Service Management Annet 2017. [Accepted]
2017 Sofiane Lagraa, Véronique Legrand, Marine Minier. Behavioral change-based anomaly detection in computer networks using data mining . International Symposium on Recent Advances in Intrusion Detection, [To be Submitted]
2016 Sofiane Lagraa, Hamida Seba. An efficient exact algorithm for triangle listing in large graphs. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery ECML-PKDD 2016
2014 Sofiane Lagraa, Alexandre Termier, and Frédéric Pétrot. Scalability bottlenecks discovery in mpsoc platforms using data mining on simulation traces. In Design Automation Conference in Europe DATE , pages 1–6, 2014
2013 Sofiane Lagraa, Alexandre Termier, and Frédéric Pétrot. Data mining mpsoc simulation traces to identify concurrent memory access patterns. In Design Automation Conference in Europe DATE , pages 755–760, 2013
2012 Sofiane Lagraa, Alexandre Termier, and Frédéric Pétrot. Automatic congestion detection in mpsoc programs using data mining on simulation traces . In RSP , pages 64–70, 2012
2011 Sofiane Lagraa, Hamida Seba, and Hamamache Kheddouci. Matchmaking owl-s processes: an approach based on path signatures. In MEDES, pages 169–176, 2011
2011 Sofiane Lagraa, Hamida Seba, Riadh Khennoufa, and Hamamache Kheddouci. A graph decomposition approach to web service matchmaking. In WEBIST, pages 31–40, 2011
Journal
Year Paper
2016 Sofiane Lagraa, Hamida Seba. An efficient exact algorithm for triangle listing in large graphs. Data Mining and Knowledge Discovery DMKD 2016, pages 1-20
2014 Sofiane Lagraa, Hamida Seba, A. MBAYA R. Khennoufa, and Hamamache Kheddouci. A distance measure for large graphs based on prime graphs. In Pattern Recognition 47(9), pages 2993–3005, 2014
2012 Hamida Seba, Sofiane Lagraa, and Hamamache Kheddouci. Alliance-based clustering scheme for group key management in mobile ad hoc networks/. The Journal of Supercomputing, 61(3):481–501, 2012
Book chapter
Year Paper
2011 Hamida Seba, Sofiane Lagraa, and Hamamache Kheddouci. Web service matchmaking by subgraph matching. In WEBIST (Selected Papers)
Award
  • Best Paper Award at DATE'2014
    Authors: Sofiane Lagraa (TIMA and LIG), Alexandre Termier (LIG), Frédéric Pétrot (TIMA)
    Title: Scalability Bottlenecks Discovery in MPSoC Platforms Using Data Mining on Simulation Traces
    at EDAA/IEEE/ACM 17th Design Automation and Test in Europe (DATE'2014) Conference in Dresden, Germany
Teaching
*** I teach Hadoop MapReduce courses for Master 2 at University of Lorraine (24 hours). Courses: HDFS, MapReduce, Hue, Pig, Hive, Sqoop, Flume
*** I taught at the Computer Science departement of University Grenoble Alpes the following courses:
  • Master 1
    • Software engineering
  • Engineer School at Polytech Grenoble
    • Assembly language (Arm)
  • L3
    • PHP - MySQL
  • L2 and L1
    • Algorithmic
    • VBA
    • HTML, CSS, Javascript
    • Database
    • C/C++
    • Network
    • Linux / shell
    • Assembly language (Motorola)
I performed 420 hours of teaching.
Students
Master
Co-Supervisor of Master2 project with Hamida Seba, Speciality: Artificial Intelligence and Decision aids.
Title : Matching large graphs
Student : Elsen R.
Master
Supervisor of Master 1 project, Speciality: Web, Informatique et Connaissance (WIC)
Title : Framework for Multiple choice on smartphone and tablet
Student : Gaëtan R. & Christian S.
Project : http://www.exam-in.fr/
Master
Supervisor of Master 1 project, Speciality: Web, Informatique et Connaissance (WIC)
Title : Forum Analysis tool
Student : Guillaume L. & Jonathan P
License
Co-Supervisor of License 3 project with Christine VERDIER, Speciality: L3-LPbiostat
Title : Development of the epidemiological survey database
Student : team of 4 students.
License
Supervisor of License 3 project, Speciality: MEtiers de l’administration et de la Sécurité des Systèmes d’Information (MESSI) at IUT2.
Title : Secure messaging for doctors
Student : Florien W.
Company : SARL SOFTIN SYSTEM.
Reviewing
Reviewer for the following conferences:
    • 4P Factory e-laboratory
    • TPC member: IFIP/IEEE International Workshop on Analytics for Network and Service Management (AnNet 2017)
Co-Reviewer for the following conferences and journal:
    • Data & Knowledge Engineering (DKE) Journal special issue (2017)
    • IFIP/IEEE International Symposium on Integrated Network Management (IM 2017)
    • 50th Design Automation Conference (DAC) 2013 (x2 papers)
    • IEEE International Conference on Data Mining series (ICDM2012, ICDM 2011).
    • Knowledge Discovery and Data Mining (KDD 2012)
    • Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2012)
Contact

Last update: 10/03/2017