International Journal on Artificial Intelligence Tools

Artificial immune systems (AISs) are computational intelligence (CI) oriented methods using information based on biological immune systems. In this study, an AIS, which combines the metaphor of clonal selection with idiotypic network theories, is developed. Although they are contradictory approaches, clonal selection and idiotypic network may prove useful in designing a stochastic global optimization tool. The AIS method consists of idiotypic network selection, somatic hypermuation, receptor editing and bone marrow operators. The idiotypic network selection operator determines the number of good solutions. The somatic hypermutation and receptor editing operators comprise the searching mechanisms for the exploration of the solution space. Diversity on the population of solutions is ensured by the bone marrow operator. The performance of the proposed AIS method is tested on a set of global constrained optimization problems (GCO), comprising of four benchmark nonlinear programming problems and four generalized polynomial programming (GPP) problems, where GPP problems are nonconvex optimization problems. The best solution found by the AIS algorithm is compared with the known global optimum. Numerical results show that the proposed method converged to the global optimal solution to each tested CGO problem. Moreover, this study compares the numerical results obtained by the AIS approach with those taken from published CI approaches, such as alternative AIS methods and genetic algorithms.

This paper addresses stability issues in incremental induction of decision trees. Stability problems arise when an induction algorithm must revise a decision tree very often and oscillations between similar concepts decrease learning speed. We review a heuristic that solves this problem and subsequently employ asymptotic analysis to approximate the basic parameters related to the estimation of computational effort in incremental learning of decision trees. We then use these approximations to simplify the heuristic, we deliver insight into its amortizing behavior and argue how they can also speed-up its execution and enhance its applicability, also providing experimental evidence to support these claims.

In many situations, an explanation of the reasons behind inconsistency in an overconstrained CSP is required. This explanation can be given in terms of minimal unsatisfiable subsets (MUSes) of constraints. This paper presents algorithms for finding minimal unsatisfiable subsets (MUSes) of constraints in overconstrained CSPs with finite domains and binary constraints. The approach followed is to generate subsets in the subset space, test them for consistency and record the inconsistent subsets found. We present three algorithms as variations of this basic approach. Each algorithm generates subsets in the subset space in a different order and curtails search by employing various search pruning mechanisms. The proposed algorithms are anytime algorithms: a time limit can be set on an algorithm's search and the algorithm can be made to find a subset of MUSes. Experimental evaluation of the proposed algorithms demonstrates that they perform two to three orders of magnitude better than the existing indirect algorithms. Furthermore, the algorithms are able to find MUSes in large CSP benchmarks.

Enhancing the content and structure of a web site is a very important task which can help to maintain people visiting a web site and gain new visits (or customers). Web mining area helps to enhance a web site organization and contents using data mining algorithms. In particular we may perform Web Mining using a Self Organizing Feature Map (SOFM or SOM) it is always needed an analysis phase by experts. To help analysts to perform this phase after SOFMs' training, many post-processing techniques have been developed (component planes, labels, etc.); however, none of these techniques are useful when working in web mining for off-line enhancements of a web site. In this paper an algorithm called Reverse Cluster Analysis (RCA) will be provided. It aims to identify important web pages based on a self organizing feature map (SOFM) when performing web text mining (WTM) and web usage mining (WUM). We successfully applied this technique in a real web site to show its effectiveness. We have extended previous work performing a comparison with another unsupervised technique, administrators survey and an extended survey.

The present study describes a practical methodology for automatic space monitoring based solely on the perceived acoustic information. Our approach is based on a two stage recognition schema that detects and classifies sound events related to hazardous situations. The main objective is to identify on time abnormal events that may lead to life-threatening situations or property damage and forward detected sound events to an authorized officer for further evaluation of the cases. We consider the case where the atypical situations of screams, explosions or gunshots take place in a metro station environment. For describing the audio signals we constructed a feature vector which includes the Mel-frequency cepstral coefficients and three MPEG-7 low level descriptors. These are subsequently fed to hidden Markov models towards representing each sound category. The accuracy of the proposed method is tested under several SNR conditions.

We propose two Euclidean minimum spanning tree based clustering algorithms — one a k-constrained, and the other an unconstrained algorithm. Our k-constrained clustering algorithm produces a k-partition of a set of points for any given k. The algorithm constructs a minimum spanning tree of a set of representative points and removes edges that satisfy a predefined criterion. The process is repeated until k clusters are produced. Our unconstrained clustering algorithm partitions a point set into a group of clusters by maximally reducing the overall standard deviation of the edges in the Euclidean minimum spanning tree constructed from a given point set, without prescribing the number of clusters. We present our experimental results comparing our proposed algorithms with k-means, X-means, CURE, Chameleon, and the Expectation-Maximization (EM) algorithm on both artificial data and benchmark data from the UCI repository. We also apply our algorithms to image color clustering and compare them with the standard minimum spanning tree clustering algorithm as well as CURE, Chameleon, and X-means.

Repetitive sequences of DNA are meaningful and of great importance to human functions. Previous researchers have proposed various methods to discover repetitive sequences in DNA sequence. However, the unknown lengths for repetitive sequences are usually predicted randomly or determined by rules of thumb rather than using a systematical criterion. We propose a new algorithm based on the cumulative Fourier spectral contents of DNA sequence to identify the candidate lengths of repetitive sequences or repeats in DNA sequences. After the candidate lengths of repeats are known, one can identify the repeats and their copy numbers using an exact method. Both of the simulated and real datasets are used to illustrate the performance of the proposed algorithm. The results are also compared to two well-known methods such as Spectral Repeat Finder (SRF) and Gibbs sampler. Furthermore, we demonstrate the use of CSRF in some well-known repeats-finding methods such as SRF, Gibbs sampler, MEME.

Context-aware systems are intended for providing services adapted to the needs of people, by taking into account their state and the information related to their environment. One alternative to represent this context information resides in the use of Semantic Web ontologies. They provide a formal vocabulary which allows to easily express and share knowledge. Additionally, several types of automatic knowledge manipulation and reasoning processes become available thanks to the formal features of such ontologies. The inclusion of context information through ontologies in Collaborative Working Environments (CWEs) may bring important benefits to team work inside an organization, such as an automatic selection between different collaborative services according to the team members' preferences and their current state. This paper describes the design and implementation of a context-reasoning system which has been integrated into a CWE architecture to take advantage of context-awareness.

Polynomial mutation is widely used in evolutionary optimization algorithms as a variation operator. In previous work on the use of evolutionary algorithms for solving multi-objective problems, two versions of polynomial mutations were introduced. The first is non-highly disruptive that is not prone to local optima and the second is highly disruptive polynomial mutation. This paper looks at the two variants and proposes a dynamic version of polynomial mutation. The experimental results show that the proposed adaptive algorithm is doing well for three evolutionary multiobjective algorithms on well known multiobjective optimization problems in terms of convergence speed, generational distance and hypervolume performance metrics.

Currently most researches in process mining focus on discovering a workflow model from an entire log. In fact, the process designers may have a partially built model and their prior knowledge is also valuable information to process mining. Besides, the large volume of log data makes the process mining a time-consuming job. It is better that the process mining be done in an incremental way. There are only a few methods that can incrementally mine a model, but they have their limitations such as incapability in handling loops or being intolerant to noise. On the other hand, loop mining is a challenging problem in process mining because the repeatedly executed tasks add complexity to the search for task precedence. This paper studies the problem of handling loops in process mining and proposes an improved incremental process mining method which supports loops. Experiments in the end show the feasibility and validity of the proposed method.