Search

Intrusion detection systems are devoted to monitor a network with aims at finding and avoiding anomalous events. In particular, we focus on misuse detection systems, which are trained to identify several known types of attacks. These can be unauthorized accesses, or denial of service attacks, among others. Whenever it scans a trace of a suspicious event, it is programmed to trigger an alert and/or to block this dangerous access to the system. Depending on the security policies of the network, the administrator may seek different requirements that will have a strong dependency on the behavior of the intrusion detection system. For a given application, the cost of raising false alarms could be higher than carrying out a preventive access lock. In other scenarios, there could be a necessity of correctly identifying the exact type of cyber attack to proceed in a given way. In this paper, we propose a multi-objective evolutionary fuzzy system for the development of a system that can be trained using different metrics. By increasing the search space during the optimization of the model, more accurate solutions are expected to be obtained. Additionally, this scheme allows the final user to decide, among a broad set of solutions, which one is better suited for the current network characteristics. Our experimental results, using the well-known KDDCup'99 problem, supports the quality of this novel approach in contrast to the state-of-the-art for evolutionary fuzzy systems in intrusion detection, as well as the C4.5 decision tree. [ABSTRACT FROM AUTHOR]

Real-parameter optimisation is a prolific research line with hundreds of publications per year. There exists an impressive number of alternatives in both algorithm families and enhancements over their respective original proposals. In this work, we analyse if this growth in the number of publications is correlated with a real progress in the field. We have selected five approaches from one of the most significant journals in the field and compared them with the winner of the competition celebrated within the IEEE Congress on Evolutionary Computation 2005. We observe that not only these methods are unable to get the good results of the winner of the competition, published several years before, but that they often avoid this type of comparison. Instead, they usually compare with other approaches from the same family. We conclude that the comparison with the state-of-the-art of the field should be mandatory to promote a real progress and to prevent that the area becomes obfuscated for outsiders. [ABSTRACT FROM AUTHOR]

Fuzzy rule-based classification systems (FRBCSs) are known due to their ability to treat with low quality data and obtain good results in these scenarios. However, their application in problems with missing data are uncommon while in real-life data, information is frequently incomplete in data mining, caused by the presence of missing values in attributes. Several schemes have been studied to overcome the drawbacks produced by missing values in data mining tasks; one of the most well known is based on preprocessing, formerly known as imputation. In this work, we focus on FRBCSs considering 14 different approaches to missing attribute values treatment that are presented and analyzed. The analysis involves three different methods, in which we distinguish between Mamdani and TSK models. From the obtained results, the convenience of using imputation methods for FRBCSs with missing values is stated. The analysis suggests that each type behaves differently while the use of determined missing values imputation methods could improve the accuracy obtained for these methods. Thus, the use of particular imputation methods conditioned to the type of FRBCSs is required. [ABSTRACT FROM AUTHOR]

Population-based meta-heuristics are algorithms that can obtain very good results for complex continuous optimization problems in a reduced amount of time. These search algorithms use a population of solutions to maintain an acceptable diversity level during the process, thus their correct distribution is crucial for the search. This paper introduces a new population meta-heuristic called 'variable mesh optimization' (VMO), in which the set of nodes (potential solutions) are distributed as a mesh. This mesh is variable, because it evolves to maintain a controlled diversity (avoiding solutions too close to each other) and to guide it to the best solutions (by a mechanism of resampling from current nodes to its best neighbour). This proposal is compared with basic population-based meta-heuristics using a benchmark of multimodal continuous functions, showing that VMO is a competitive algorithm. [ABSTRACT FROM AUTHOR]

Some theoretical models have been proposed in the literature to predict dynamics of real-coded evolutionary algorithms. These models are often applied to study very simplified algorithms, simple real-coded functions or sometimes these make difficult to obtain quantitative measures related to algorithm performance. This paper, trying to reduce these simplifications to obtain a more useful model, proposes a model that describes the behavior of a slightly simplified version of the popular real-coded CHC in multi-peaked landscape functions. Our approach is based on predicting the shape of the search pattern by modeling the dynamics of clusters, which are formed by individuals of the population. This is performed in terms of dynamical probability distributions as a basis to estimate its averaged behavior. Within reasonable time, numerical experiments show that is possible to achieve accurate quantitative predictions in functions of up to 5D about performance measures such as average fitness, the best fitness reached or number of fitness function evaluations. [ABSTRACT FROM AUTHOR]

Multiobjective genetic fuzzy rule selection is based on the generation of a set of candidate fuzzy classification rules using a preestablished granularity or multiple fuzzy partitions with different granularities for each attribute. Then, a multiobjective evolutionary algorithm is applied to perform fuzzy rule selection. Since using multiple granularities for the same attribute has been sometimes pointed out as to involve a potential interpretability loss, a mechanism to specify appropriate single granularities at the rule extraction stage has been proposed to avoid it but maintaining or even improving the classification performance. In this work, we perform a statistical study on this proposal and we extend it by combining the single granularity-based approach with a lateral tuning of the membership functions, i.e., complete contexts learning. In this way, we analyze in depth the importance of determining the appropriate contexts for learning fuzzy classifiers. To this end, we will compare the single granularity-based approach with the use of multiple granularities with and without tuning. The results show that the performance of the obtained classifiers can be even improved by obtaining the appropriate variable contexts, i.e., appropriate granularities and membership function parameters. [ABSTRACT FROM AUTHOR]

Nowadays, large scale optimisation problems arise as a very interesting field of research, because they appear in many real-world problems (bio-computing, data mining, etc.). Thus, scalability becomes an essential requirement for modern optimisation algorithms. In a previous work, we presented memetic algorithms based on local search chains. Local search chain concerns the idea that, at one stage, the local search operator may continue the operation of a previous invocation, starting from the final configuration reached by this one. Using this technique, it was presented a memetic algorithm, MA-CMA-Chains, using the CMA-ES algorithm as its local search component. This proposal obtained very good results for continuous optimisation problems, in particular with medium-size (with up to dimension 50). Unfortunately, CMA-ES scalability is restricted by several costly operations, thus MA-CMA-Chains could not be successfully applied to large scale problems. In this article we study the scalability of memetic algorithms based on local search chains, creating memetic algorithms with different local search methods and comparing them, considering both the error values and the processing cost. We also propose a variation of Solis Wets method, that we call Subgrouping Solis Wets algorithm. This local search method explores, at each step of the algorithm, only a random subset of the variables. This subset changes after a certain number of evaluations. Finally, we propose a new memetic algorithm based on local search chains for high dimensionality, MA-SSW-Chains, using the Subgrouping Solis Wets' algorithm as its local search method. This algorithm is compared with MA-CMA-Chains and different reference algorithms, and it is shown that the proposal is fairly scalable and it is statistically very competitive for high-dimensional problems. [ABSTRACT FROM AUTHOR]

In the classification framework there are problems in which the number of examples per class is not equitably distributed, formerly known as imbalanced data sets. This situation is a handicap when trying to identify the minority classes, as the learning algorithms are not usually adapted to such characteristics. An usual approach to deal with the problem of imbalanced data sets is the use of a preprocessing step. In this paper we analyze the usefulness of the data complexity measures in order to evaluate the behavior of undersampling and oversampling methods. Two classical learning methods, C4.5 and PART, are considered over a wide range of imbalanced data sets built from real data. Specifically, oversampling techniques and an evolutionary undersampling one have been selected for the study. We extract behavior patterns from the results in the data complexity space defined by the measures, coding them as intervals. Then, we derive rules from the intervals that describe both good or bad behaviors of C4.5 and PART for the different preprocessing approaches, thus obtaining a complete characterization of the data sets and the differences between the oversampling and undersampling results. [ABSTRACT FROM AUTHOR]

An introduction to the journal is presented in which the editor discusses an article on genetic fuzzy data mining, an article on Mamdani fuzzy systems, and the application of subgroup discovery methods for analysis of patients and the arrival time in Psychiatric Emergency Department.