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.
New trends in neural computation, now dealing with distributed learning on pervasive sensor networks and multiple sources of big data, make necessary the use of computationally efficient techniques to be implemented on simple and cheap hardware architectures. In this paper, a nonuniform quantization at the input layer of neural networks is introduced, in order to optimize their implementation on hardware...
We discuss several ways to accelerate genetic algorithm-based instance selection, where the two objectives are a minimal number of training instances and maximal accuracy of the classifier (we use neural networks) on the test data. We discuss several ways to accelerate the process, but we especially focus on two parameters: fitness function and chromosome length reduction. We evaluate different fitness...
The recently proposed trainable COSFIRE filters are highly effective in a wide range of computer vision applications, including object recognition, image classification, contour detection and retinal vessel segmentation. A COSFIRE filter is selective for a collection of contour parts in a certain spatial arrangement. These contour parts and their spatial arrangement are determined in an automatic...
Web 2.0 enabled users to share their experiences, views, and opinions. One of the key products of Web 2.0 is Twitter, a social media site with hundreds of millions of users. These users tweet whatever they want to share with other people. The aim of this paper is to classify the tweets into subjective and objective tweets. We group words people use in Twitter into objective and subjective words, creating...
We address the issue of image recognition from the position of performance and robustness. Lately, impressive advancements have been made in this field, largely due to the substantial increase in computational power available at the consumer price. However, there still remains a need to optimize classification algorithms while keeping them sensibly robust to various space-time variations. We propose...
In this study, a two level options trading strategy is modelled and optimized with Genetic Algorithms and Particle Swarm Optimization for profit maximization. In the first level, the trend is found and in the second level, options trading strategies for the particular trend are determined. The strike prices and expiration dates of the traded options are optimized and tested on 5 different Exchange...
Nowadays many researches related to classifier design are trying to exploit strength of the ensemble learning. Such hybrid approach looks for the valuable combination of individual classifiers’ outputs, which should at least outperforms quality of the each available individuals. Therefore the classifier ensembles are recently the focus of intense research. Basically, it faces with two main problems...
Traditional classifiers in steganalysis are no longer applicable when faced with massive feature set and high-dimensional sample set. We proposed a kind of selective ensemble classifier for universal steganalysis based on virus-evolutionary genetic algorithm. After generating some base learners, we selected some of them according to genetic optimization with an additonal virus population. The final...
Semi-supervised learning combines labeled and unlabeled examples in order to find better future predictions. Usually, in this area of research we have massive amounts of unlabeled instances and few labeled ones. In this paper each instance has attributes from multiple sources of information (views) and a genetic algorithm is applied for regression function learning. Based on the few labeled examples...
This paper deals with the optimization of the multiuser (MU) pilot design for orthogonal-frequency-division-multiple-access (OFDMA) systems perturbed by the I/Q imbalance (IQI) impairment. For the frequency-selective Rayleigh propagation channels, the proposed approach exploits the decoupled direct and image channels, and the knowledge of the effective channel statistics to realize dual linear-minimum-mean-square-error...
Using the counter propagation artificial neural network (CPANN) to diagnose the DGA fault, network structural parameters should be set, such as the training epochs, network size etc. When user to set, it would be affect by the artificial subjective factors. If we use the traversal search way, it would be the consumption of computing and time. So this article employed parallel genetic algorithm to...
The features are considered the cornerstone of text summarization. The most important issue is what feature to be considered in a text summarization process. Including all the features in the summarization process may not be considered as an optimal solution. Therefore, other methods need to be deployed. In this paper, random five features used and investigated using a (pseudo) Genetic concept as...
Citrus quality classification is an important and widely studied topic since it has significant role in its market price determination. Due to citrus quality indicators series nonlinearity and no-stationary, the accuracy of conventional mostly used methods including linear discriminant analysis, K-means clustering and neural network has been limited. The use of support vector machine (SVM) has been...
Aiming at the defects of BP neural network, analyzed the disadvantages of the genetic algorithm and the common quantum genetic algorithm. Used the diversity of population and the rapidity of convergence of the real-number coded double-chain quantum genetic algorithm which was combined with BP neural network to modify the weights and thresholds of the neural network, and the modified neural network...
In order to solve the fault diagnosis problem of vibration Parameter, this dissertation proposes the application of adaptive neural network-based fuzzy inference system to engine error diagnosis. Different from the fuzzy inference system, the membership function adopted in this method is no longer a fixed entity but an optimal one achieved by the practice of neural network, which adopts the method...
Adaptive neural network-based fuzzy inference system (ANFIS) was applied to build a fault diagnosis model of automobile engine, the thesis, with the construction of ANFIS, by using gradient descent genetic algorithm and optimization of system parameters of neutral network learning algorithm, inputs the fusion data into ANFIS, the ANFIS fault diagnosis model adopts the method of information fusion...
This paper, in order to reduce fault and improve ratio of recognition, build adaptive neural network-based fuzzy inference system (ANFIS), which was applied to build a fault diagnosis model of automobile engine, adopts the method of information fusion in entropy method to optimize the input interface. To reduce the impact of excessive parameters on classification accuracy and cost, it also raises...
In response to the ongoing discussion on how electric power distribution systems should evolve under the Smart Grid Initiative, an optimization problem is defined to simultaneously determine optimal locations for Distributed Generation (DG) and feeder interties in a legacy radial distribution system to improve reliability in the islanded mode of operation. For that purpose, an evolutionary approach...
In order to deal with the defects of the poor convergences and easily immerging in partial minimum frequently, a new algorithm is proposed based on the combination of genetic algorithm and BP neural network, which is called GA- BP algorithm. This algorithm is applied to optimization of initial weights of BP Network, the structure and learn rule. It searches through the total solution space and can...
In this paper, we present the problem of appropriate feature selection for constructing a Maximum Entropy (ME) based Named Entity Recognition (NER) system under the multiobjective optimization (MOO) framework. Two conflicting objective functions are simultaneously optimized using the search capability of MOO. These objectives are (i). the dimensionality of features, which is tried to be minimized,...
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.