The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
With AdaBoost being constructed, base classifiers are more and more concentrating on instances which are difficult to classify. And base classifiers are only favourite to these instances. After a base classifier is constructed, it's voting weight for final decision is determined and be the same to all test instances no matter which class a test instance belongs to. Considering these problems, the...
One of the important issues in the design of fuzzy classifier is the formation of fuzzy if-then rules and the membership functions. This paper presents a Niched Pareto Genetic Algorithm (NPGA) approach to obtain the optimal rule-set and the membership function. To develop the fuzzy system the rule set and the membership functions are encoded into the chromosome and evolved simultaneously using NPGA...
An adaptive k-nearest neighbor algorithm (AdaNN) is brought forward in this paper to overcome the limitation of the traditional k-nearest neighbor algorithm (kNN) which usually identifies the same number of nearest neighbors for each test example. It is known that the value of k has crucial influence on the performance of the kNN algorithm, and our improved kNN algorithm focuses on finding out the...
PSO has been proved as an effective supervised learning system in recent years, but it's not an effective method for incremental learning problems. Aiming at the incremental learning target for classification, a hybrid algorithm of Particle Swarm Optimization (PSO) and Artificial Immune System (AIS) called Immune based PSO (IPSO) is presented in this paper. IPSO inherits the incremental learning ability...
We explore the framework of permutation-based p-values for assessing the behavior of the classification error. In this paper we study two simple permutation tests. The first test estimates the null distribution by permuting the labels in the data; this has been used extensively in classification problems in computational biology. The second test produces permutations of the features within classes,...
In fuzzy rule based classification systems, a high number of predictive attributes leads to an explosion of the number of generated rules and can affect the learning algorithm precision. Thus, the increase of the number of features can degrade the predictive capacity of the fuzzy rule based classification systems. In this article, we propose a supervised learning method by automatic generation of...
Many classification algorithms use the concept of distance or similarity between patterns. Previous work has shown that it is advantageous to optimize general Euclidean distances (GED). In this paper, we optimize data transformations, which is equivalent to searching for GEDs, but can be applied to any learning algorithm, even if it does not use distances explicitly. Two optimization techniques have...
This paper revisits supervised machine learning for multiclass problems with the assumption that all classes cannot be represented in a training set. This is common in many applications in which there are numerous classes or in which some classes are exceedingly rare. In this paper we propose the use of a decision function to serve in place of the decision boundaries which are used in many machine...
In the d-dimensional feature space, the classification weight is defined against the different contribution of every feature that used to classification on the training sample set. And the classification weight calculates the membership functions which set up unascertained classification. Then a novel supervised clustering algorithm based on above is given. The algorithm is concise in calculation,...
In conventional RBF Network structure, different layers perform different tasks. Hence, it is useful to split the optimization process of hidden layer and output layer of the network accordingly. This study proposes hybrid learning of RBF Network with Particle Swarm Optimization (PSO) for better convergence, error rates and classification results. The hybrid learning of RBF Network involves two phases...
Manifold learning is one of the efficient nonlinear dimensionality reduction techniques, which can be used to fault feature extraction. But they are not taking the class information of the data into account. In this paper, a new supervised Laplacian eigenmaps algorithm (S-LapEig) for classification is proposed first. Via utilizing class information to guide the procedure of nonlinear mapping, the...
Fast fuzzy support vector machines (FFSVMs) based on the convex hulls are proposed in this paper. Firstly, the convex hull of each class data is generated by using the quick hull algorithm, and the data points lying inside the convex hull are not important to form FSVMs and then discarded. Secondly, the reduced training set consisting of the convex points is used to train the FFSVMs. Thirdly, the...
This paper presents an approach for designing classifiers for a multiclass problem using Genetic Programming (GP). The proposed approach takes an integrated view of all classes when GP evolves. An individual of the population will be represented using multiple trees. The GP is trained with a set of N training samples in steps. A concept of unfitness of a tree is used in order to improve genetic evolution...
Ensemble of classifiers has been an interesting research topic in the area of machine learning. In this paper, we propose a new ensemble scheme which focuses on driving the relationship between multiple learning algorithms and variant data distributions. The advantage of the framework can form an expressive hypotheses combination allowing a set of learning algorithms with respect to the data distributions,...
Like most classification techniques, the existing support vector machines (SVM) approaches are challenged to correctly classify their input when the data points are either very close to the decision boundary or very dissimilar from the training data set. In both situations, most classifiers including SVMs will still give a prediction by assigning the test point to one of the classes. However, when...
In this paper, a fuzzy pattern classification tuning approach is proposed, which is based on fusion concept. In this method, tuning parameters are learned in a training procedure, enabling system to be capable of managing individual classification task. Fuzzy c-means, as a specific instance of Tuning Reference, is employed as a tool to offer membership function which is used for making decisions and...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.