Abstract: How to find and detect novel or unknown network attacks is one of the most important objectives in current intrusion detection systems. In this paper, a rule evolution approach based on Genetic Programming (GP) for detecting novel attacks on networks is presented and four genetic operators, namely reproduction, mutation, crossover, and dropping condition operators, are used to evolve new rules. New rules are used to detect novel or known network attacks. A training and testing dataset proposed by DARPA is used to evolve and evaluate these new rules. The proof of concept implementation shows that a rule generated by GP has a low false positive rate (FPR), a low false negative rate and a high rate of detecting unknown attacks. Moreover, the rule base composed of new rules has high detection rate with low FPR. An alternative to the DARPA evaluation approach is also investigated.

Abstract: In recent years, there has been a large amount of disparate work concerning the representation and reasoning with qualitative preferential information by means of approaches to nonmonotonic reasoning. Given the variety of underlying systems, assumptions, motivations, and intuitions, it is difficult to compare or relate one approach with another. Here, we present an overview and classification for approaches to dealing with preference. A set of criteria for classifying approaches is given, followed by a set of desiderata that an approach might be expected to satisfy. A comprehensive set of approaches is subsequently given and classified with respect to these sets of underlying principles.

Abstract: Face recognition has recently received much attention as a biometric security means well accepted by users. 3D face recognition, exploiting the geometry of the facial surface, addresses the two most sensitive aspects of face recognition: viewpoint and illumination changes. This paper presents a 3D face recognition approach based on the geometrical comparison of a set of profiles. These profiles are obtained as parallel planar cuts of the facial surfaces. In a first step, normalised profiles are extracted from each face independently thanks to the intrinsic symmetry of faces. The homologue profiles of test and reference surfaces are then matched two by two. The individual distances of profile pairs are combined into a global distance. The global distance after optimisation is used as criterion to state the similarity of faces. Results are presented for a population of 81 persons of a 3D face database of intermediate quality.

Abstract: The design of Boolean functions with properties of cryptographic significance is a hard task. In this paper, we adopt an unorthodox approach to the design of such functions. Our search space is the set of functions that possess the required properties. It is “Boolean-ness” that is evolved.

Abstract: A system that combines real-time face tracking as well as the localization of facial landmarks in order to improve the authenticity of fingerprint recognition is introduced. The intended purpose of this application is to assist in securing public areas and individuals, in addition to enforce that the collected sensor data in a multi modal person authentication system originate from present persons, i.e. the system is not under a so called play back attack. Facial features are extracted with the help of Gabor filters and classified by SVM experts. For real-time performance, selected points from a retinotopic grid are used to form regional face models. Additionally only a subset of the Gabor decomposition is used for different face regions. The second modality presented is texture-based fingerprint recognition, exploiting linear symmetry. Experimental results on the proposed system are presented.

Abstract: In e-business, disputes between two or more parties arise for various reasons and involve different issues. Thus, resolution of these disputes frequently relies on some form of negotiation. This article presents a general problem-solving framework for modeling multi-issue multilateral agent negotiation using fuzzy constraints in e-business. Fuzzy constraints are thus used not only to define each agent's demands involving human concepts, but also to represent the relationships among agents. A concession strategy, based on fuzzy constraint-based problem-solving, is proposed to relax demands, and a trade-off strategy is presented to evaluate existing alternatives. This approach provides a systematic method for reaching an agreement that benefits all agents with a high satisfaction degree of constraints. Meanwhile, by applying the method, agents can move toward an agreement more quickly, because their search focuses only on the feasible solution space. An example application to negotiate an insurance policy among agents is provided to demonstrate the usefulness and effectiveness of the proposed framework.

Abstract: Although ceteris paribus preference statements concisely represent one natural class of preferences over outcomes or goals, many applications of such preferences require numeric utility function representations to achieve computational efficiency. We provide algorithms, complete for finite universes of binary features, for converting a set of qualitative ceteris paribus preferences into quantitative utility functions.

Abstract: The emerging standards for the specification of Web Services support the publication of the static interfaces of the operations they may execute. However, little attention is paid to the management of long-lasting interactions between the service providers and their consumers. Although this is not an issue in the case of “one-shot” services, it challenges the provision of services requiring the exchange of multiple messages between the business partners. In this article, we present a conversational model supporting the management of long-lasting interactions where several messages have to be exchanged before the service is completed. Our model aims at facilitating the consumers during the service invocation because in this way the establishment of short-term business relations can be simplified. To this extent, we provide a computational framework that can be exploited to manage a conversation between the consumer and the service provider. Our framework is inspired from the research developed in Computational Linguistics and in the area of Multi-Agent Systems to manage human-to-computer and agent-to-agent dialog. However, we employ techniques suitable to comply with the emerging Web Service standards and with the scalability requirements of the Internet.

Abstract: Reference LIA-ARTICLE-2004-002View record in Web of Science Record created on 2006-12-13, modified on 2017-05-12

Abstract: The contribution of this work is designing and developing enhanced market-driven agents with the flexibility to (1) respond to changing market conditions, and (2) raise and relax trade aspirations. Previous theoretical analyses have shown that market-driven agents (MDAs) make prudent compromises by reacting to changing market situations by taking into account factors such as competition, deadlines, and trading options. This work augments the design of an MDA with three fuzzy decision controllers that guide the agent in (i) relaxing trade aspiration in face of intense negotiation pressure, and (ii) raising trade aspiration in extremely favorable markets. Results from extensive simulations conducted using an implemented testbed suggest that when compared to MDAs, agents in this work achieved (1) higher success rates in reaching deals, (2) higher average utilities, and (3) higher expected utility.

Abstract: Airline luggage contains a wide variety of objects and their automated image analysis require good quality image segmentation. Given the fact that such images are highly cluttered, it is non trivial task to optimise image segmentation algorithms. In this paper we present a methodology for optimising image segmentation algorithms based on image properties without manual intervention. The methodology computes image properties such as average edge gradient strength, inter- vs. intra-cluster distances using image colour features, and colour purity of resultant regions, to train a neural network that maps these to ground-truth labelling on the acceptability (good or bad) of the solution (resultant segmentation). We show that on unseen test data, this methodology performs extremely well by correctly predicting the optimal parameters of image segmentation algorithms used.

Abstract: In symmetric cryptology the resistance to attacks depends critically on the nonlinearity properties of the Boolean functions describing cipher components like Substitution boxes (S-boxes). Some of the most effective methods known to generate functions that satisfy multiple criteria are based on evolutionary heuristics. In this paper, we improve on these algorithms by employing an adaptive strategy. Additionally, using recent improvements in the understanding of these combinatorial structures, we discover essential properties of the graph formed by affine equivalence classes of Boolean functions, which offers several advantages as a conceptual model for multiobjective seeking evolutionary heuristics. Finally, we propose the first major global cooperative effort to discover new bounds for cryptographic properties of Boolean functions.

Abstract: Techniques taken from the field of Artificial Intelligence, especially Evolutionary Computation (Genetic Algorithms and Genetic Programming, but also others) are steadily becoming more and more present in the area of computer security, both in network/host security and in the very demanding area of cryptology. In recent years, many algorithms that take advantage of approaches based on Evolutionary Computation have been proposed, for example, in the design and analysis of a number of new cryptographic primitives, ranging from pseudo-random number generators to block ciphers, in the cryptanalysis of state-of-the-art cryptosystems, and in the detection of network attacking patterns, to name a few. There is a growing interest from the computer security community toward Evolutionary Computation techniques, as a result of these recent successes, but there still are a number of open problems in the field that should be addressed.

Abstract: Most of the research in the area of expert finding focuses on creating and maintaining centralized directories of experts' profiles, which users can search on demand. However, in a distributed multiagent-based software environment, the autonomous agents are free to develop expert models or model fragments for their own purposes and from their viewpoints. Therefore, the focus of expert finding is shifting from the collection at one place as much data about a expert as possible to accessing on demand from various agents whatever user information is available at the moment and interpreting it for a particular purpose. This paper outlines purpose-based expert modeling as an approach for finding an expert in a multiagent portfolio management system in which autonomous agents develop expert agent models independently and do not adhere to a common representation scheme. This approach aims to develop taxonomy of purposes that define a variety of context-dependent user modeling processes, which are used by the users' personal agents to find appropriate expert agents to advise users on investing strategies.

Abstract: A simple way of creating new and very efficient distinguishers for cryptographic primitives, such as block ciphers or hash functions, is introduced. This technique is then successfully applied over reduced round versions of the block cipher TEA, which is proven to be weak with less than five cycles.

Abstract: Recent research has shown that adaptive neural based control systems are very effective in restoring stability and control of an aircraft in the presence of damage or failures. The application of an adaptive neural network with a flight critical control system requires a thorough and proven process to ensure safe and proper flight operation. Unique testing tools have been developed as part of a process to perform verification and validation (V&V) of real time adaptive neural networks used in recent adaptive flight control system, to evaluate the performance of the on line trained neural networks. The tools will help in certification from FAA and will help in the successful deployment of neural network based adaptive controllers in safety-critical applications. The process to perform verification and validation is evaluated against a typical neural adaptive controller and the results are discussed.

Abstract: Security protocols play an important role in modern communications. However, security protocol development is a delicate task, and experience shows that computer security protocols are notoriously difficult to get right. Recently, Clark and Jacob provided a framework for automatic protocol generation based on combinatorial optimization techniques and the symmetric key part of BAN logic. This paper shows how such an approach can be further developed to encompass the full BAN logic without the loss of efficiency and thereby synthesize public key protocols and hybrid protocols.

Abstract: The paper describes a new multiagent system with enhanced capabilities obtained through a hybrid of intelligent techniques. The processing in the model is handled by two types of agents: distributed agents and a central administrator agent. Localized processing at the individual agents is carried out using mathematical techniques and genetic algorithms. The central administrator agent dynamically obtains information about the problem domain from the Internet and maintains a knowledge pool using a clustering technique called the growing self-organizing map (GSOM). Distributed agents communicate with the central administrator agent if they need further knowledge about the problem domain to provide solutions to user-defined tasks. The approach integrates traditional mathematical, data mining, and evolutionary techniques with a multiagent system. The proposed system is implemented as a travel optimizer application for the e-tourism domain. Finally, the possibilities of integrating the proposed technique with currently available e-tourism applications to provide the customer with enhanced solutions are identified.

Abstract: Antiterrorist analysts often need to work in teams with the requirement to analyze voluminous amounts of dynamic information in order to assess potential terrorist threats. Analysts have a high cognitive demand complicated by factors that typically the information has restrict access and requires special expertise for interpretation. The goal of this research is to enhance team performance by modeling and implementing a cognitive agent architecture capable of proactively seeking, linking and sharing information using knowledge and experience distributed among team members. The agent architecture is empowered by a collaborative RPD model - a novel team-based naturalistic decision making process derived from Klein's recognition-primed decision framework.

Abstract: Many organizations and governments have the need to monitor areas for intrusions and, once detected, to identify the type of potential intruder(s) present. Applications include perimeter security at installations such as airports and critical infrastructure, as well as military situation awareness in monitoring demilitarized zones, or other areas where activity of interest may occur. Seismic signal detectors can be used in many of these applications. Time-frequency response (TFR) signals are generated and must be classified as being generated by particular targets of interest. Experiments were conducted using real data collected at Marine Corps Base, Camp Pendleton, California, USA. Seismic signal detectors were used to monitor signals generated by individual people, groups of people, and vehicles of different types. Evolutionary computation was combined with neural networks to analyze the TFR signals and classify the acquired data. The results indicated the practical application of classifying signals based on their seismic signature.

Abstract: This study presents a Montgomery multiplication architecture that uses an irreducible all one polynomial (AOP) in GF(2m) based on a programmable cellular automata (PCA). The proposed architecture has the advantage of high regularity and a reduced latency based on combining the characteristics of the irreducible AOP and PCA. The proposed architecture can be used to implement modular exponentiation, division, and inversion architectures.

Abstract: Biometric systems are designed by expert developers who look - with trial-and-error approaches - for reasonable solutions by considering the available hardware architecture, some - possibly conflicting - quality goals, and the application constrains. Typically drawbacks of these approaches are waste of time and results far from the optimum. We propose a new design methodology for multimodal biometric systems that applies high-level system design techniques to better structure the design procedure. The proposed methodology avoids the drawbacks of the common design practice and allows to create a flexible general-purpose and effective design environment for multimodal biometric systems.

Abstract: Trading in the financial markets often requires that information be available in real time to be effectively processed. Furthermore, complete information is not always available about the reliability of data, or its timeliness—nevertheless, a decision must still be made about whether to trade or not. We propose a mechanism whereby different data sources are monitored, using Semantic Web facilities, by different agents, which communicate among each other to determine the presence of good trading opportunities. When a trading opportunity presents itself, the human traders are notified to determine whether or not to execute the trade. The Semantic Web, Web Services, and URML technologies are used to enable this mechanism. The human traders are notified of the trade at the optimal time so as not to either waste their resources or lose a good trading opportunity. We also have designed a rudimentary prototype system for simulating the interaction between the intelligent agents and the human beings, and show some results through experiments on this simulation for trading of the Chicago Board Options Exchange (CBOE) options.

Abstract: Most existing formalizations treat belief change as a single-step process, and ignore several problems that become important when a theory, or belief state, is revised over several steps. This paper identifies these problems, and argues for the need to retain all of the multiple possible outcomes of a belief change step, and for a framework in which the effects of a belief change step persist as long as is consistently possible. To demonstrate that such a formalization is indeed possible, we develop a framework, which uses the language of PJ-default logic (Delgrande and Jackson 1991) to represent a belief state, and which enables the effects of a belief change step to persist by propagating belief constraints. Belief change in this framework maps one belief state to another, where each belief state is a collection of theories given by the set of extensions of the PJ-default theory representing that belief state. Belief constraints do not need to be separately recorded; they are encoded as clearly identifiable components of a PJ-default theory. The framework meets the requirements for iterated belief change that we identify and satisfies most of the AGM postulates (Alchourron, Gardenfors, and Makinson 1985) as well.

Abstract: In this paper we describe a face detection algorithm, the key idea being to determine roughly the skin regions of a 2D color image and searching for eyes through them. The technique is based on a support vector machine trained to separate sub images representing eyes from others. The algorithm can be used in face image database management systems both as a first step of a person identification, and to discriminate the images on the basis of the number of faces in them.

Abstract: We propose a human behaviour recognition system based on video sequences. Our aim is to identify one among several kinds of actions performed by a single person in a particular scenery. Each frame will be processed, detecting the moving objects and using a new statistical-based algorithm to erase shadows. The final step consists of the extraction of different kinds of other human contour points. Correspondences between each contour pattern and posture will be achieved by using classical pattern classifiers (K-neighbours and Mahalanobis distance), and also with a modified self organized map (SOM). We will analyze and compare the obtained results, combining the contour patterns with some information concerning temporal relationship in consecutive movements, in order to improve the correct action detection.

Abstract: This paper concerns a new technology for synchronised management of knowledge, traffic and resources in a global network of integrated operational decision support (IODS) systems for airport and airline operators, pilots and air traffic controllers The paper introduces the development of the IODS global processes and control mechanisms in learning air space-time control structures and clearance-categories of four dimensional end-to-end trajectories of flights It introduces a global neural network of control of air space time allocation and of traffic flows through satellites A patent application is pending regarding IODS global processes and control mechanisms introduced in this paper

Abstract: This paper shows how action theories, expressed in an extended version of the language , can be naturally encoded using Prioritized Default Theory. We also show how prioritized default theory can be extended to express preferences between rules. This extension provides a natural framework to introduce different types of preferences in action theories—preferences between actions and preferences between final states. In particular, we demonstrate how these preferences can be expressed within extended prioritized default theory. We also discuss how this framework can be implemented in terms of answer set programming.

Abstract: Adaptive business agents operate in electronic marketplaces, learning from past experiences to make effective decisions on behalf of their users. How best to design these agents is an open question. In this article, we present an approach for the design of adaptive business agents that uses a combination of reinforcement learning and reputation modeling. In particular, we take into account the fact that multiple selling agents may offer the same good with different qualities, and that selling agents may alter the quality of their goods. We also consider the possibility of dishonest agents in the marketplace. Our buying agents exploit the reputation of selling agents to avoid interaction with the disreputable ones, and therefore to reduce the risk of purchasing low value goods. We then experimentally compare the performance of our agents with those designed using a recursive modeling approach. We are able to show that agents designed according to our algorithms achieve better performance in terms of satisfaction and computational time and as such are well suited for the design of electronic marketplaces.