Catálogo de publicaciones - libros

Classical Rough Sets Theory (CRST) is thought to be an effective mathematical approach to discovering rules from Decision Table (DT). Every entry in DT must be unique and certain qualitative value of attribute. However, there are always heterogeneous entries in DT from complex decision problem, that is, entries with a continuous quantitative attribute, unknown entries or multi-valued entries, and these types of entries often occur in same DT. The DT with these entries named Continuous Uncertain Decision Table (CUDT) cannot be analyzed directly by CRST. Fortunately, by modeling those three types of entries in CUDT with Fuzzy Sets theory (FST), we found that CUDT can be transformed into a special DT called Extended Decision Table (EDT) in which each entry is associated with a membership degree. An extended CRST is proposed to transform the CUDT into EDT and to calculate the lower approximations and the boundaries of decision concepts in EDT.

To solve the intrusion detection question, we introduce the fuzzy logic into Organization CoEvolutionary algorithm and present the algorithm of Organization CoEvolutionary Fuzzy Classification (OCEFC). In this paper, we give an intrusion detection models based on OCEFC. After illustrating our model and applying it to the real-world network datasets KDD Cup 1999, we obtain the better performance than other traditional methods.

In data mining for knowledge explanation purposes, we would like to build simple transparent fuzzy models. Compared to other fuzzy models, simple fuzzy logic rules (IF ... THEN... rules) based on triangular or trapezoidal shape fuzzy sets are much simpler and easier to understand. For fuzzy rule based learning algorithms, choosing the right combination of attributes and fuzzy sets which have the most information is the key point to obtain good accuracy. On the other hand, the fuzzy ID3 algorithm gives an efficient model to select the right combinations. We therefore discover the set of simple fuzzy logic rules from a fuzzy decision tree based on the same simple shaped fuzzy partition, after dropping those rules whose credibility is less than a reasonable threshold, only if the accuracy of the training set using these rules is reasonably close to the accuracy using fuzzy decision tree. The set of simple fuzzy logic rules satisfied with this condition is also able to be used to interpret the information of the tree. Furthermore, we use the fuzzy set operator “OR” to merge simple fuzzy logic rules to reduce the number of rules.

The is a new type of information system developed from the incomplete information system by introducing the new concept of based on the possible relations among the domains of the attributes in the information system. In this paper, we will discuss the knowledge reduction in the information system based on sub-consciousness, we also propose the concept of rationally guided emotional reduction in the information system based on sub-consciousness which is then compared with the rational reduction and the emotional reduction in the information system based on sub-consciousness.

Conic-cylindrical gear reduction unit as a high-performance power transmission device is widely used to build various machineries. There are lots of fuzzy factors in its manufacturing process and operation environment, which should be taken into consideration in the design process. Fuzzy physical programming is an effective multiobjective optimization method which incorporates fuzziness in its problem formulation. The fuzzy physical programming model for the optimal design of two-stage conic-cylindrical gear reduction unit is developed in this paper, and genetic algorithm is used to solve the model. An example is given to illustrate that fuzzy physical programming can consider the fuzziness of conic-cylindrical gear reduction unit substantially, and conforms more perfectly to the engineering realities.

In this paper we present NLOMJ—a natural language object model in Java with English as the experiment language. It describes the grammar elements of any permissible expression in a natural language and their complicated relations with each other with the concept “Object” in OOP. Directly mapped to the syntax and semantics of the natural language, it can be used in information retrieval as a linguistic method. Around the UML diagram of the NLOMJ the important classes (Sentence, Clause and Phrase) and their sub classes are introduced and their syntactic and semantic meanings are explained.

Reconstruction of fingerprint ridge lines is a critical pre-processing step in the identification of poor quality fingerprint images. This paper presents a new fingerprint ridge line reconstruction approach by way of ridge line tracing. In our research, the fingerprint ridge line in a gray scale image is viewed as a track of a ridge segment moving along the ridge. The curve tracing problem is solved by the target tracking technique in computer vision. We first formulate the model of fingerprint ridge line segments and then apply a target tracking method to trace each of the ridge lines. In addition, a feedback technique is adopted to correct the fingerprint directional image in each tracing step in order to improve tracing accuracy. By connecting all the traced ridge line segments, a polyline reconstruction of the ridge line can be obtained. We objectively assess the performance of this approach by using NIST fingerprint images.

Data Mining is the process of posing queries and extracting useful information, patterns and trends previously unknown from large quantities of data. Agents are defined as software entities that perform some set of tasks on behalf of users with some degree of autonomy. This research work deals about developing a automated data mining system which encompasses the familiar data mining algorithms using intelligent agents in object oriented databases and proposing a framework. Because the data mining system uses the intelligent agents, a new user will be able to interact with the data mining system without much data mining technical knowledge. This system will automatically select the appropriate data mining technique and select the necessary fields needed from the database in a right time without expecting the users to specify the specific technique and the parameters. Also a new framework is proposed for incorporating intelligent agents with automated data mining. One of the major goals in developing this system is to give the control to the computer for learning automatically by using intelligent agents for the exploratory data mining.

A Bayesian optimization algorithm (BOA) for unmanned aerial vehicle (UAV) path planning is presented, which involves choosing path representation and designing appropriate metric to measure the quality of the constructed network. Unlike our previous work in which genetic algorithm (GA) was used to implement implicit learning, the learning in the proposed algorithm is explicit, and the BOA is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. Experimental results demonstrate that this approach can overcome some drawbacks of other path planning algorithms. It is also suggested that the learning mechanism in the proposed approach might be suitable for other multivariate encoding problems.

Associative classification has aroused significant attention in recent years. This paper proposed a novel interestingness measure, named dilated chi-square, to statistically reveal the interdependence between the antecedents and the consequent of classification rules. Using dilated chi-square, instead of confidence, as the primary ranking criterion for rules under the framework of popular CBA algorithm, the adapted algorithm presented in this paper can empirically generate more accurate and much more compact decision lists.