Research

Goal: To develop a software system for opinion mining by analyzing various techniques and key tasks involved in opinion mining from online resources.

Members: Dalila Bouras (PhD student), Supervisors: H. Bendjenna, M. Amroune.

Starting Date: Completed Project

Full Description: With the growing availability and popularity of opinion-rich resources such as online review, news reports, client emails, sites and personal blogs, new opportunities and challenges arise as people now can, and do, actively use information technologies to seek out and understand the opinions of others. In this project we survey and analyze various techniques that have been developed for the key tasks of opinion mining. We wish to develop a software system for opinion mining on the basis of our survey and analysis.

Goal: To develop a Vehicular Cloud Networking system by combining the benefits of VANETs and cloud computing for improving traffic safety and efficiency.

Members: Dr. Sahraoui Abdelatif, Dr. Achouri Mounir, Pr. Makhlouf Derdour.

Starting Date: Ongoing Project

Full Description: One such network that has received a lot of interest in the last couple of years is the Vehicular Ad-Hoc Network (VANET). VANET has become an active area of research, standardization, and development because it has tremendous potential to improve vehicle and road safety, traffic efficiency, and convenience as well as comfort to both drivers and passengers. On the other hand, cloud computing is supposed to be the next big thing because of its scalability, PaaS, IaaS, SaaS and other important characteristics. In this project, we envision a new vehicular networking system, Vehicular Cloud Networking.

Goal: To develop a model that integrates palmprint biometrics with cryptographic schemes to reduce fraud risks in e-banking transactions and enhance customer trust.

Members: Pr. A. Meraoumia, Dr. O. Chergui, Pr. H. Bendjenna.

Starting Date: Ongoing Project

Full Description: Today, the Internet plays a big and increasingly important role in our modern societies. The financial institution, such the bank, is one of these societies whose started using internet as a distribution channels for the services they offer. The large diffusion of e-banking service requires more stringent methodologies to identify customers in order to prevent any fraud incident. The objective of this project is to develop a design model that implements the concepts of fuzzy commitment scheme based on palmprint biometrics to reduce risks of fraud and to improve customer’s trust in online banking transactions. Thus, to enhance the discriminating capability of the feature vector, extracted from each palmprint modality, several feature extraction methods are used. After that, the financial transaction information is encrypted using a random key (public or private) and then transmitted to the bank institution. In the bank server, an efficiency scheme for the key retrieval is implemented in order to determine the cryptographic key (private key) which used for decrypt the message and then treated. To evaluate our scheme a multispectral palmprint database of 400 is used.