Government supported organizations like Energetic Bear are targeting these critical infrastructures for espionage purpose. This is possible by attacking industrial control systems (ICS). The goal of this project is to analyze Energetic Bear malware and to suggest mitigation strategies.

This project is a step towards a research solution for Service Level Agreement (SLA) Management in cloud. We will explore current solutions of SLA Management, which will umbrella Resource Management and Scheduling in Cloud Computing. This project will deliver a formal model for SLA Management which will map Service Level Objects (SLOs) to optimization methods generically. Which will be applicable to various kinds of cloud setups and deliver robust SLA management.

A cyber security monitoring system is a comprehensive system that is usually deployed by the organizations to safeguard themselves from potential risk and threats prevailing and attacking the cyber environment and its assets. The system monitors different devices like FWs, IDS and devices etc. to collect information about the unusual activity in the cyber environment. Logging it, identifying it and alerting the concerned entities of the possible danger.

The heart of any data diode is unidirectional communication there are different ways of implementing this link; best of all is the hardware implementation i.e using optical hardware devices. Firewalls can be configured to restrict two ways communication but they are prone to errors, and thus data diodes are the ultimate solution for the system that requires data to only flow out.

Cloud computing in education is seen as the next wave of information technology, cloud-based education have become the impetus of innovating teaching model, learning style and learning environment. The world is exploring new instruction model on cloud-based education to promote effective learning actively. Cloud-based applications and services are available to many students who own computer, smartphone, and mobile devices such as iPad. Cloud based software such as Google Docs, SkyDrive, Evernote, and many other educational applications made it possible for everyone to learn whatever and whenever they want. It will be a Proof of Concept at a campus. The PoC will be limited in scope and the purpose will be to demonstrate the feasibility of this idea.

The aim of this project s to design a SNMP based cyber security monitoring system. In this there are two main components, first is the application running on the number of devices such routers, switches, servers and desktops etc. which are connected to the network and whom we will want to monitor. Second is the administrative application called as Network Monitoring System (NMS) or manager, through which we will monitor and manage the network attached devices.

This project aims at automating mitotic counting and improving their detection accuracy. A dataset of histopathology slides comprised of samples from all the three grades will be used for training and testing. Usually, samples for each grade are not available in equal proportion therefore such dataset are sometimes severely imbalanced which makes the task of training the classifier hard.

The objective of this project is to develop a software suite for multiple instance machine learning. Multiple instance learning is a generalization of supervised learning in which the objective is to learn a concept given “bags” of training examples. A positive labeled bag can have both positive and negative examples of the concept in it whereas a negative bag only has negative examples. Such machine learning techniques are finding application in a variety of interesting areas such as Bioinformatics, Computer Vision and Computer Security.

Project Description