Catálogo de publicaciones - libros

In this paper, we present a novel rule evaluation support method for post-processing of mined results with rule evaluation models based on objective indices. Post-processing of mined results is one of the key issues to make a data mining process successfully. However, it is difficult for human experts to evaluate many thousands of rules from a large dataset with noises completely. To reduce the costs of rule evaluation procedures, we have developed the rule evaluation support method with rule evaluation models, which are obtained with objective indices of mined classification rules and evaluations of a human expert for each rule. To evaluate performances of learning algorithms for constructing rule evaluation models, we have done a case study on the meningitis data mining as an actual problem. In addition, we have also evaluated our method on four rulesets from the four UCI datasets. Then we show the availability of our rule evaluation support method.

The poor characterisation of contaminated soils is likely to result in high costs, restricted choice in landfill disposal sites and future environmental impact. This makes that big quantities of soils are still waiting to be recovered. Until now the tools used to detect contaminated soil have been very generic, without any criteria of prioritization. Usually, simulation studies are used to classify contaminated soils, however these systems need a large quantity of data that is difficult to obtain and manage, which means that the results obtained are subject to large uncertainty. Recently, Artificial Intelligence techniques have been used to tackle this problem. In this work we propose the use of Fuzzy Expert Systems to classify the soils. Classical decision rules have shown to be interpretable, efficient, problem independent and able to treat large scale applications, but they are also recognised as highly unstable classifiers with respect to minor perturbations in the training data. In our problem, the data is subject to uncertainty, for this reason we propose the fuzzyfication of the variables. In the study many different variables have been taken into account: Physic and Chemical characteristics of the soil and pollutants, toxicological properties, and environment and social conditions. After applying Fuzzy Expert Systems at different levels, we obtain a ranking of the soils according to their risk of contamination. The results have been contrasted with another Multicriteria Decision Making system.

The main goal of agent modelling is to extract and represent the knowledge about the behaviour of other agents. Nowadays, modelling an agent in multi-agent systems is increasingly becoming more complex and significant. Also, robotic soccer domain is an interesting environment where agent modelling can be used. In this paper, we present an approach to classify and compare the behaviour of a multi-agent system using a coach in the soccer simulation domain of the RoboCup.

The whole software industry has an awful footpath for delivering software on-time and on-budget. Probably, one reason is coming from not deal with the imperfection of information when they use a lot of human process. In this paper, we propose the use of fuzzy measures in contrast with crisp measures of traditional models and, therefore, apply of appropriate aggregators. Traditional models of software cost estimation are constructed from project databases and they describe cost drivers in terms of linguistic estimations using vague terms like “low” or “high”, and such expressions are also used in obtaining actual predictions. But cost drivers are in many cases abstract concepts that are better estimated by breaking them down in a number of second-level aspects. The method proposed is based both, on a concrete study of the use of linguistic variable human categorizations and, on level aspects that are defined by layer and are easy to raise using appropriate aggregators. Moreover, the proposed scheme can have different planes according to the model morphology.

Several researchers on country-of-origin (coo) have expressed their interest in knowing how consumers’ emotional reactions toward coo-cues affect product attitude formation. This paper shows how Fuzzy Set Theory might serve as a useful approach to that problem. Data was gathered by means of self-administered questionnaires. Technically, orness of OWA-operators enabled us to distinguish consumers expressing highly positive versus less positive emotions toward coo. It appeared that this variance in emotional estate goes together with a difference in aggregating product-attribute beliefs.

Non-monotonic fuzzy measures induced by an intuitinistic fuzzy set are introduced. Then, using the Choquet integral with respect to the non-monotonic fuzzy measure, the weighted distance between two intuitionistic fuzzy sets is defined. As it will be shown here, under some conditions, the weighted distance coincides with the Hamming distance.

An evaluation method of fuzzy numbers is presented from the viewpoint of aggregation operators in decision making modeling. The method is given by the quasi-arithmetic means induced from weighted aggregation operators with a decision maker’s subjective utility. The properties of the weighted quasi-arithmetic mean and its translation invariance are investigated. For the mean induced from the dual aggregation operators, a formula for the calculation is also given. The movement of the weighted quasi-arithmetic means is studied in comparison between two decision maker’s utilities, which are essentially related to their attitude in decision making. Several examples are examined to discuss the properties of this defuzzification method.

Yager [1] introduced several families of ordered weighted averaging (OWA) operators, in which the associated weights depend on the aggregated arguments. In this paper, we develop a new dependent OWA operator, and study some of its desirable properties. The prominent characteristic of this dependent OWA operator is that it can relieve the influence of unfair arguments on the aggregated results. Finally, we give an example to illustrate the developed operator.

Agent technology provides a new way to model many complex problems like financial investment planning. With this observation in mind, a financial investment planning system was developed from agent perspectives with 12 different agents integrated. Some of the agents have similar problem solving and decision making capabilities. The results from these agents require to be combined. Ordered Weighted Averaging (OWA) operator was chosen to aggregate different results. Details on how OWA was applied as well as appropriate evaluation are presented.

The concepts of generated universal fuzzy measures and of basic generated universal fuzzy measures are introduced. Special classes and properties of generated universal fuzzy measures are discussed, especially the additive, the symmetric and the maxitive case. Additive (symmetric) basic universal fuzzy measures are shown to correspond to the Yager quantifier-based approach to additive (symmetric) fuzzy measures. The corresponding Choquet integral-based aggregation operators are then generated weighted means (generated OWA operators).