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.
In reinforcement learning, it is important to get nearly right answers early. Good prediction early can reduce the prediction error afterward and accelerate learning speed. We propose fuzzy Q-map, function approximation algorithm based on on-line fuzzy clustering in order to accelerate learning. Fuzzy Q-map can handle the uncertainty owing to the absence of environment model. Applying membership function...
Clustering ensembles have emerged as a powerful method for improving both the robustness and the stability of unsupervised classification solutions. However, finding a consensus clustering from multiple clusterings is a difficult problem that can be approached from graph-based, combinatorial or statistical perspectives. A consensus scheme via the genetic algorithm based on information theory is proposed...
To the problem that it is hard to determine the clustering number and the abnormal points by using the clustering validity function, an effective clustering partition model based on the genetic algorithm is built in this paper. The solution to the problem is formed by the combination of the clustering partition and the encoding samples, and the fitness function is defined by the distances among and...
In this paper, we propose a method to find the anomalous behaviors in network traffic. We map the network connection records into different feature spaces typically of high dimension according to their protocols and services. In training, we perform clustering to group training data points into clusters, from which we select some clusters as normal and known-attack profile according to a simple, but...
Adaptive polyclonal algorithm is the improved one of clonal selection algorithm, and its convergence speed is much faster. This paper intends to direct a novel clustering analysis by means of the affinity function that the adaptive polyclonal clustering strategy affects. The clustering algorithm has the advantage that it does not depend on priori knowledge and has nothing to do data distribution,...
The rival penalized expectation-maximization (RPEM) algorithm has demonstrated its powerful capability to perform the model selection automatically in the context of mixture model. However, the performance may be degraded when irrelevant variables are included. To overcome this drawback, we adopt the concept of feature salience as the feature weight to measure the relevance to the clusters in the...
Many applications require the management of spatial data in very large spatial databases. We propose a grid-based clustering algorithm to process spatial data. This algorithm uses a grid-based method to identify data that are compressible and data that must be maintained in memory. Thus the algorithm compresses data without decreasing the quality of clustering. Through one scan over a database, our...
In this paper, we explore a new problem of simultaneously mining diagnostic genes and specific phenotypes from microarray data using unsupervised method. A novel type of cluster called LC-Cluster is proposed to address this problem. The idea behind the solution is motivated by recent biological discovery and origins from current bicluster model or emerging pattern, but differs substantially from either...
Cluster validity index is used to evaluate the clustering result yielded by the fuzzy clustering algorithm. In this paper, a new cluster validity index is proposed to determine the optimal fuzzy c-partition produced by the fuzzy c-means algorithm. The proposed index introduces two evaluation factors: distribution density and uncertainty. The first factor measures the extent of closeness or compactness...
Pattern-based clustering is widely applied in bioinformatics and biomedical Recently, mining high quality pattern-based clusters has become an important research direction. However, the existing methods were neither efficient in large data set nor precise at measuring the quality of clusters. These problems have greatly limited the methods' application in large data set. This paper proposes a new...
In this paper, we propose techniques converting odor information into digital data to provide representation, digital coding, and clustering of various odors for olfactory information communication. About forty emotional adjectives describing various kinds of odors are selected as expressional receptors. Each of the odors is expressed by a definite adjective. The odors expressed using emotional receptors...
In order to raise Chinese pills' productive efficiency, a microwave resonator (MR) with a center hole is improved for measuring density and moisture content of Chinese pill material. When pill material passes through the center hole, its density and moisture will cause resonant frequency excursion (RFE) and microwave attenuation (MA). Under the same moisture, density is only related to MA. If RFE...
In this paper, we propose a particle swarm based algorithm to cluster peer-to-peer network hosts. Previously, the clustering of network hosts are mainly according to their connectivity [?], according to the RTTs between the hosts by probing each other, or cluster randomly. In our work, the information used to cluster the network hosts are getting from the network positioning system. The algorithm...
With the increasing number of Web documents in the Internet, the most popular keyword-matching-based search engines, such as Google, often return a long list of search results ranked based on their relevance and importance to the query. To cluster the search engine results can help users find the results in several clustered collections, so it is easy to locate the valuable search results that the...
Synthetic aperture radar (SAR) image classification is becoming increasingly important in military or scientific research. SAR image classification based on unsupervised learning usually requires optimization of some metrics. Local optimization techniques frequently fail because functions of these metrics with respect to transformation parameters are generally nonconvex and irregular and, therefore,...
Fuzzy c-means (FCM) clustering algorithm has been widely used in automated image segmentation. However, the standard FCM algorithm is sensitive to noise because of taking no into account the gray and spatial information of pixel. The paper proposes an improved FCM algorithm for image segmentation. We use the degree of gray similarity and distribution statistics of the neighbor pixels to form a new...
This paper is to investigate a new unsupervised approach for the extracted objects based on synthetic aperture radar (SAR) image using improving fuzzy clustering method. The traditional fuzzy c-means clustering (FCM) is very sensitive to the initial value and the number of clusters. The accurate initial value and number of clusters are important parameters to get the accurate result in FCM. SAR image...
In underdetermined blind separation, the number of sensors is less than that of source signals, and it is well known that source signals can be recovered through the two-step algorithms generally. But people often suppose that the number of source signals is known when they estimate the mixture matrix by the k-mean clustering algorithm. In fact, the number of source signals is unknown or blind, so...
The wavelet transform modulus extremum is considered as one of the most meaningful characteristics of a signal. This paper proposes a feature based on the density of modulus extrema of the wavelet frame representation for texture classification. It is compared with existing features by using three representative classifiers, k-nearest neighbor classifier, learning vector quantization and support vector...
Documents may suffer from perspective distortion when captured with hand-held digital cameras. A morphology based rectification method is proposed to recover fronto-parallel views of perspectively distorted documents images. Firstly we extend the nearest-neighbor (NN) clustering technique in document skew rectification to locate the horizontal vanishing point of the text plane. Secondly we partition...
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.