Aller au contenu. | Aller à la navigation

Outils personnels


Voir le monde en interaction

  • logo-enssib.jpg
Vous êtes ici : Accueil / Actualités / PHD position: "Anomaly detection in dynamic attributed networks with community structure." (LabHC, Saint-Etienne)

PHD position: "Anomaly detection in dynamic attributed networks with community structure." (LabHC, Saint-Etienne)

PHD position at Hubert Curien Laboratory Saint-Etienne

Anomaly detection in dynamic attributed networks with community structure.

Laboratory: Laboratoire Hubert Curien - University of Saint Etienne

Christine Largeron 
Baptiste Jeudy

In the recent few years, network data has become ubiquitous and has attracted great interest in the data mining community. A variety of methods and software solutions have been proposed to ease their analysis. On the other hand, anomaly detection is an important problem in many application domains.
Thus anomaly detection in graphs, and in particular in dynamic graphs (graphs which change over time) is a hot topic and the subject of recent research papers [Ranshous2015, Eberle2007, Noble2003, Akoglu2014, Gupta2014].
Anomaly detection in dynamic networks deals with the problem of finding nodes, edges, points in time or substructures that are dissimilar with respect to the rest of the network. Applications includes discovery of extreme physical events (cyclones) [Chen2013], intrusion detection [Ding2012], communication networks analysis [Priebe2005],

In this thesis, we propose to study anomaly detection in dynamic attributed networks with community structure and to focus on detecting anomalous nodes. We propose to do this using both the relationship between nodes (the graph) and the attributes. More precisely, a node could be considered as anomalous if it presents atypical attribute values given its community membership. This kind of approach has never been tried to our knowledge. 

Working environment
The PhD candidate will work at Laboratoire Hubert Curien (University of Saint Etienne, in the Data Intelligence team. This team has an expertise in social networks, deep learning, pattern mining and anomaly detection among others.
The two supervisors for this PhD thesis are:
* Christine Largeron, Professor in Computer Science since 2006. Her current research focuses on social mining and text mining. 
* Baptiste Jeudy, Associate Professor in Computer Science since September 2006.  His research topics are data mining, in particular in graphs and graphs sequences.

Funding the Ph.D. fellowship is funded for 3 years and is monthly funded about approximatively 1450 €.

 Profile of the candidate
The candidate should have a master degree or equivalent in computer science. The subject is at the intersection of several domains: graph theory (applied to social network and community detection), anomaly detection (statistics and machine learning) and big data (the considered networks can be huge). Thus the candidate should have strong backgrounds in several of these topics.
Other required skills :

  • Good abilities in algorithm design and programming ;
  • a very good level (written and oral) in English ;
  • good communication skills (oral and written);
  • autonomy and motivation for research.

Application instructions
Send your application with a CV, your last grade certificate (if you are currently finishing your Master’s degree, we need an official list of the grades you obtained so far in this degree with your rank among your peers), some recommendation letters and a specific motivation letter to and
The application is opened until the 24th April. Some interviews will be offered between the 25th April and the 4th May. 
The final decision will be given in June. The PhD thesis is expected to start in September (or October) 2018.