Abstract The problem of intrusion detection is studied extensively in computer security. The development of security mechanisms, like Intrusion Detection Systems, is of great importance in order to preserve the confidentiality, integrity and availability of data stored in computers. Off-line intrusion detection can be accomplished by searching audit trail logs of user activities for matches to patterns of events required for... The problem of intrusion detection is studied extensively in computer security. The development of security mechanisms, like Intrusion Detection Systems, is of great importance in order to preserve the confidentiality, integrity and availability of data stored in computers. Off-line intrusion detection can be accomplished by searching audit trail logs of user activities for matches to patterns of events required for known attacks. This is a combinatorial optimization problem, the NPcomplete. Metaheuristics offer an alternative to solve this type of problems as databases of events and attacks grow. This paper presents an approach using an evolutionary algorithm, Harmony Search metaheuristic, to detect intrusions. Experiments are performed to show its effectiveness to detect the unseen intrusion attacks with high detection rate and recognize normal network traffic with low false alarm rate. Furthermore, comparisons with a “Biogeography” inspired intrusion detection approach are made. The results confirm the good behavior of our new approach. Read More Read Less

Abstract A neural network intrusion detection system (IDS) can be effective against network attacks. However, their effectiveness can be reduced by changes in the neural network architecture. One problem is determining the number of hidden layer neurons. This can lead to reduced detection and high failure rates. This paper describes the affects of architecture on the performance of IDSs while finding... A neural network intrusion detection system (IDS) can be effective against network attacks. However, their effectiveness can be reduced by changes in the neural network architecture. One problem is determining the number of hidden layer neurons. This can lead to reduced detection and high failure rates. This paper describes the affects of architecture on the performance of IDSs while finding a means to choose the proper architecture and number of hidden neurons. This method reduces the need for trial and error in determining the number of hidden layer neurons in a multi layer feed forward neural network IDS. Read More Read Less

Abstract Authentication is a mechanism to establish proof of identities. Common practise for authentication is to challenge user to present authentication credential. User who can present the valid credential is considered as authenticated identity. Traditional authentication systems have a static security requirement which does not consider the change of user behavior or environment. In contrast, adaptive authentication systems are risk-based authentication... Authentication is a mechanism to establish proof of identities. Common practise for authentication is to challenge user to present authentication credential. User who can present the valid credential is considered as authenticated identity. Traditional authentication systems have a static security requirement which does not consider the change of user behavior or environment. In contrast, adaptive authentication systems are risk-based authentication that consider those changes to identify high-risk and suspicious illegitimate login attempts. In this paper, we examine some of the adaptive authentication systems that have been proposed recently in the literature. Some of the issues and challenges when developing an adaptive authentication system are also discussed. We also introduce our own Unified Authentication Platform (UAP) which incorporates adaptive control. We describe two typical processes that are used in our Adaptive UAP. Read More Read Less

Abstract While several models are provided in the literature to estimate different features of projects management: duration, effort, cost, and investment, they are not tailored to the cost estimation of security projects. We provide in this paper a new model for the managerial aware estimation of the effort required to conduct a security project. This model takes into consideration the effort... While several models are provided in the literature to estimate different features of projects management: duration, effort, cost, and investment, they are not tailored to the cost estimation of security projects. We provide in this paper a new model for the managerial aware estimation of the effort required to conduct a security project. This model takes into consideration the effort related to monitoring, awareness, decision making, organizational, and decision support. Three effort estimation models are proposed depending on the information system size, the complexity of existing security policy, and the enterprise size. In the proposed model, the cost associated to the achievement of a security project is deduced from the estimated effort. Read More Read Less

Abstract Trust is a dynamic concept that evolves during a relationship. We study the initiation phase of trust toward an unknown website. We propose three sets of drives of initial trust: relational concepts (perceived value and satisfaction), in individual concept (trust propensity) and a contextual concept (structural assurance). We adopted the quasi- experimentation method for data collection. We have obtained 388... Trust is a dynamic concept that evolves during a relationship. We study the initiation phase of trust toward an unknown website. We propose three sets of drives of initial trust: relational concepts (perceived value and satisfaction), in individual concept (trust propensity) and a contextual concept (structural assurance). We adopted the quasi- experimentation method for data collection. We have obtained 388 questionnaires. We have used structural equation modeling and we adopted multi-group analyses to test the moderator role of trust propensity. Read More Read Less