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.
Visually Impaired Malaysians find great difficulty in recognizing bank notes as there are no commercially available aid for her currency and not much research done in the area. A very low cost, high efficient system has been proposed to help visually impaired Malaysians identify her bank notes. This system is built around the unique color of her bank note. The system features a sound module, color...
A novel feature selection approach is proposed for data space defined over continuous features, which obtains a subset of features,such that it can discriminate class labels of objects and its discriminant ability is not inferior to that of the original features,so to effectively improve the learning performance and intelligibility of the classification model. According to the spatial distribution...
Many of the network data visualization tools or applications are designed and being applied in network data visualization system which are particularly for users with advanced network knowledge even though the tools are indispensable by diverse computer users. In this paper, we proposed and presented an adaptive statistical analysis learning approach that is able to adapt to the user feedback after...
With the growing diverse demands for Internet applications, network security issues become more acute. To address the appropriate network security from network intrusion detection event has become an important research in network security. In this paper, user behavior features are extracted to create the model for the user transmission behavior. The demands for anomaly detection and the specific characteristics...
This paper proposes a novel method to generate labels for grouping and organizing the search results returned by auxiliary search engines. It has applied statistical techniques to measure the quantities of co-occurrence keywords for forming the label matrix of them, and then agglomerated them into higher-level clusters by clustering algorithm in order to classify the results which return from the...
When there is no sufficient labeled instances, supervised dimensionality reduction methods tend to perform poorly due to overfitting. In such cases, unlabeled instances are used to improve the performance. In this paper, we propose a dimensionality reduction method called semi-supervised TransductIve Discriminant Analysis (TIDA) which preserves the global and geometrical structure of the unlabeled...
Naïve Bayes classifier is proved to be one of the most effective classifier an be used widely. It applies statistical theory to text classification. This paper researched and implemented a Chinese text classifier using JAVA base on Naïve Bayes Method. First of all, this paper described test classification system, the content includes text information expressing, extracting and the method of Chinese...
The main function of load profiles is to provide information on the consumer's electrical consumption. Such information is useful for both consumers and suppliers. Consumers can participate in the retail market and keep track of their actual power consumption and suppliers could use such information for load management, develop electricity tariffs, etc. Main objective of this paper is to do cluster...
In this paper, a new scalability of hybrid fuzzy clustering algorithm that incorporates the Fuzzy C-means into the Quantum-behaved Particle Swarm Optimization algorithm is proposed. The QPSO has less parameters and higher convergent capability of the global optimizing than Particle Swarm Optimization algorithm. So the iteration algorithm is replaced by the new hybrid algorithm based on the gradient...
For improving the detection efficiency of hidden information blind detection system, an improved hidden information detection method based rough set theory is proposed against the high dimension of statistical features and high relevance about images. First, an improved general steganalysis system framework is proposed with practical method and steps; second, the Algorithm based on the rough set theory...
Vision-based vehicle make and model recognition is a hot topic in the domain of intelligent transportation systems. But it is difficult to recognize the exact make and model of a vehicle due to the influence of some factors, for example, the view variations. In this paper, we present a new method for vehicle make and model recognition with variant views. We take Gabor wavelet coefficients as our initial...
Recently traffic identification based on Machine Learning (ML) techniques has attracted a great deal of interest. Two challenging issues for these methods are how to deal with encrypted flows and cope with the rapid growing number of new application types correctly and early. We propose a hybrid traffic identification method and a novel unsupervised clustering algorithm, On-Line Density Based Spatial...
Web mining focuses on extracting useful information from large volumes of Web data. Web usage mining (WUM) is one of important application which applies Web mining techniques to discovery usage patterns from Web accessing data. Meanwhile clustering performs a key role in distinguishing different kinds of usage patterns from raw data. Considering usage features of activities, information scope and...
This article dynamically analyzes large scale XML documents of experts' instances according to thought of particle swarm and algorithm of dynamical cluster. It can quickly find out the same or similar experts' instances in Long-distance Medical System, so it improved the efficiency and accuracy of the querying, and provided us a new thinking for future instance studying in Long-distance Medical System.
Text Categorization (TC) is an important component in many information organization and information management tasks. In many TC applications, the case-base grows at a fast rate and this causes inefficiency in the case retrieval process. Using Case-Base Maintenance learning via the GC (Generalization Capability) algorithm, which can reduce the case number into KNN algorithm, can improve efficiency...
As an important technique in modern sociology, social network analysis has gained a lot of attention from many disciplines, and been used as important complements to traditional statistics and data analysis. In order to make it affordable for analysts with massive and fast growing networks, we present X-RIME, a cloud-based library for large scale social network analysis. We propose an implementation-oriented...
Selective ensemble learning is a learning algorithm, trains a number of based classifier and selects some of them to ensemble. Through the selective ensemble, the algorithm would be more effective than each single one and better than the algorithm that select all the based classifier, and the algorithm would have effective generalization ability. In this paper, we apply a multi-lever selective ensemble...
We present a method for automatic algorithm recognition, which consists of two phases. First, the target algorithms are converted into characteristic vectors, which are computed based on static analysis of program code including various statistics of language constructs and analysis of Roles of Variables. In the second phase, the algorithms are classified based on these vectors using the C4.5 decision...
Many statistic based machine learning methods depend on the estimation of probability density function from observations. Non-parametric density estimation algorithms based on minimizing expirical risk using support vector machine (SVM) are quite general and powerful, but have a significant disadvantage in the smoothness of estimation result. In this paper, we studies the vicinal risk minimization...
In this paper, an improved Differential Evolution algorithm (ACDE-O) with cluster number oscillation for automatic crisp clustering has been presented. The proposed algorithm needs no prior knowledge of the number of clusters of the data. Rather, it finds the optimal number of clusters on the processing with stable and fast convergence, cluster number oscillation mechanism will search more possible...
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.