MCBÜ Açıkerişim Sistemi :: Browsing by Subject "Learning algorithms"

Browsing by Subject "Learning algorithms"

Now showing 1 - 20 of 42

Improved approach to the solution of inverse kinematics problems for robot manipulators
(Elsevier Science Ltd, 2000) Karlik B.; Aydin S.
A structured artificial neural-network (ANN) approach has been proposed here to control the motion of a robot manipulator. Many neural-network models use threshold units with sigmoid transfer functions and gradient descent-type learning rules. The learning equations used are those of the backpropagation algorithm. In this work, the solution of the kinematics of a six-degrees-of-freedom robot manipulator is implemented by using ANN. Work has been undertaken to find the best ANN configurations for this problem. Both the placement and orientation angles of a robot manipulator are used to fin the inverse kinematics solutions.
Power factor correction technique based on artificial neural networks
(2006) Sagiroglu S.; Colak I.; Bayindir R.
This paper presents a novel technique based on artificial neural networks (ANNs) to correct the line power factor with variable loads. A synchronous motor controlled by the neural compensator was used to handle the reactive power of the system. The ANN compensator was trained with the extended delta-bar-delta learning algorithm. The parameters of the ANN were then inserted into a PIC 16F877 controller to get a better and faster compensation. The results have shown that the proposed novel technique developed in this work overcomes the problems occurring in conventional compensators including over or under compensation, time delay and step changes of reactive power and provides accurate, low cost and fast compensation compared to the technique with capacitor groups. © 2006 Elsevier Ltd. All rights reserved.
An intelligent power factor corrector for power system using artificial neural networks
(2009) Bayindir R.; Sagiroglu S.; Colak I.
An intelligent power factor correction approach based on artificial neural networks (ANN) is introduced. Four learning algorithms, backpropagation (BP), delta-bar-delta (DBD), extended delta-bar-delta (EDBD) and directed random search (DRS), were used to train the ANNs. The best test results obtained from the ANN compensators trained with the four learning algorithms were first achieved. The parameters belonging to each neural compensator obtained from an off-line training were then inserted into a microcontroller for on-line usage. The results have shown that the selected intelligent compensators developed in this work might overcome the problems occurred in the literature providing accurate, simple and low-cost solution for compensation. © 2008 Elsevier B.V. All rights reserved.
Design optimization of moment resisting steel frames using a cuckoo search algorithm
(Civil-Comp Press, 2012) Saka M.P.; Doǧan E.
The Cuckoo search algorithm is a recent addition to metaheuristic techniques. It simulates the breeding behaviour of certain cuckoo species into a numerical optimization technique. Cuckoo birds lay their eggs in the nests of other host birds so that their chicks when hatched can be nurtured by the host birds. The optimum design algorithm presented for moment resisting steel frames is based on the cuckoo search algorithm. The design algorithm selects the appropriate W sections for the beams and column of a steel frame out of 272 W sections listed in the LRFD-AISC (Load and Resistance Factor Design, American Institute of Steel Construction) [52] such that the code requirements are satisfied and the weight of steel frame is the minimum. Code specifications necessitate the consideration of a combined strength constraint with lateral torsional buckling for beam-column members. Furthermore displacement constraints as well as inter-storey drift restrictions of multi-storey frames are also included in the design formulation. Further constraints related with the constructability of a steel frame are also considered. The number of steel frames are designed by the algorithm presented to demonstrate its efficiency. The same steel frames are also designed by using the big bang-big crunch algorithm as well as the particle swarm optimizer for comparison. © Civil-Comp Press, 2012.
Hunting search algorithm based design optimization of steel cellular beams
(2013) Doǧan E.
The present study examines a hunting search based optimum design algorithm for cellular beams. Hunting search is a numerical optimization method inspired by group hunting of animals. The proposed algorithm selects the optimum UB section to be used in the production of a cellular beam subjected to a general loading, the optimum holes diameter and number of these holes in the beam. Furthermore, this selection is also carried out such that the design limitations are satisfied and the weight of the cellular beam is the minimum. A design example is considered to demonstrate the application of the optimum design algorithm developed.
Solving design optimization problems via hunting search algorithm with Levy flights
(Techno-Press, 2014) Doʇan E.
This study presents a hunting search based optimum design algorithm for engineering optimization problems. Hunting search algorithm is an optimum design method inspired by group hunting of animals such as wolves, lions, and dolphins. Each of these hunters employs hunting in a different way. However, they are common in that all of them search for a prey in a group. Hunters encircle the prey and the ring of siege is tightened gradually until it is caught. Hunting search algorithm is employed for the automation of optimum design process, during which the design variables are selected for the minimum objective function value controlled by the design restrictions. Three different examples, namely welded beam, cellular beam and moment resisting steel frame are selected as numerical design problems and solved for the optimum solution. Each example differs in the following ways: Unlike welded beam design problem having continuous design variables, steel frame and cellular beam design problems include discrete design variables. Moreover, while the cellular beam is designed under the provisions of BS 5960, LRFD-AISC (Load and Resistant Factor Design-American Institute of Steel Construction) is considered for the formulation of moment resisting steel frame. Levy Flights is adapted to the simple hunting search algorithm for better search. For comparison, same design examples are also solved by using some other well-known search methods in the literature. Results reveal that hunting search shows good performance in finding optimum solutions for each design problem. Copyright © 2014 Techno-Press, Ltd.
Artificial immune system based Web page classification
(Springer Verlag, 2015) Onan A.
Automated classification of web pages is an important research direction in web mining, which aims to construct a classification model that can classify new instances based on labeled web documents. Machine learning algorithms are adapted to textual classification problems, including web document classification. Artificial immune systems are a branch of computational intelligence inspired by biological immune systems which is utilized to solve a variety of computational problems, including classification. This paper examines the effectiveness and suitability of artificial immune system based approaches for web page classification. Hence, two artificial immune system based classification algorithms, namely Immunos-1 and Immunos-99 algorithms are compared to two standard machine learning techniques, namely C4.5 decision tree classifier and Naïve Bayes classification. The algorithms are experimentally evaluated on 50 data sets obtained from DMOZ (Open Directory Project). The experimental results indicate that artificial immune based systems achieve higher predictive performance for web page classification. © Springer International Publishing Switzerland 2015.
Backtracking search algorithm-based optimal power flow with valve point effect and prohibited zones
(Springer Verlag, 2015) Kılıç U.
This paper presents a solution technique for optimal power flow (OPF) with valve-point effect and prohibited operating zones of power systems using backtracking search algorithm (BSA). BSA is a new population-based evolutionary algorithm. The most important property of the algorithm is not over precision to initial of value, unlike many other heuristic algorithms. The proposed algorithm having four different cases is tested on IEEE-30 bus test system. The results of BSA are compared to those reported in literature. Thus, its validity for so applications in this area is proved. In this paper, OPF problem of power systems is solved by BSA for the first time. © 2014, Springer-Verlag Berlin Heidelberg.
A fuzzy-rough nearest neighbor classifier combined with consistency-based subset evaluation and instance selection for automated diagnosis of breast cancer
(Elsevier Ltd, 2015) Onan A.
Breast cancer is one of the most common and deadly cancer for women. Early diagnosis and treatment of breast cancer can enhance the outcome of the patients. The development of classification models with high accuracy is an essential task in medical informatics. Machine learning algorithms have been widely employed to build robust and efficient classification models. In this paper, we present a hybrid intelligent classification model for breast cancer diagnosis. The proposed classification model consists of three phases: instance selection, feature selection and classification. In instance selection, the fuzzy-rough instance selection method based on weak gamma evaluator is utilized to remove useless or erroneous instances. In feature selection, the consistency-based feature selection method is used in conjunction with a re-ranking algorithm, owing to its efficiency in searching the possible enumerations in the search space. In the classification phase of the model, the fuzzy-rough nearest neighbor algorithm is utilized. Since this classifier does not require the optimal value for K neighbors and has richer class confidence values, this approach is utilized for the classification task. To test the efficacy of the proposed classification model we used the Wisconsin Breast Cancer Dataset (WBCD). The performance is evaluated using classification accuracy, sensitivity, specificity, F-measure, area under curve, and Kappa statistics. The obtained classification accuracy of 99.7151% is a very promising result compared to the existing works in this area reporting the results for the same data set. © 2015 Elsevier Ltd. All rights reserved.
On the performance of ensemble learning for automated diagnosis of breast cancer
(Springer Verlag, 2015) Onan A.
The automated diagnosis of diseases with high accuracy rate is one of the most crucial problems in medical informatics. Machine learning algorithms are widely utilized for automatic detection of illnesses. Breast cancer is one of the most common cancer types in females and the second most common cause of death from cancer in females. Hence, developing an efficient classifier for automated diagnosis of breast cancer is essential to improve the chance of diagnosing the disease at the earlier stages and treating it more properly. Ensemble learning is a branch of machine learning that seeks to use multiple learning algorithms so that better predictive performance acquired. Ensemble learning is a promising field for improving the performance of base classifiers. This paper is concerned with the comparative assessment of the performance of six popular ensemble methods (Bagging, Dagging, Ada Boost, Multi Boost, Decorate, and Random Subspace) based on fourteen base learners (Bayes Net, FURIA, Knearest Neighbors, C4.5, RIPPER, Kernel Logistic Regression, K-star, Logistic Regression, Multilayer Perceptron, Naïve Bayes, Random Forest, Simple Cart, Support Vector Machine, and LMT) for automatic detection of breast cancer. The empirical results indicate that ensemble learning can improve the predictive performance of base learners on medical domain. The best results for comparative experiments are acquired with Random Subspace ensemble method. The experiments show that ensemble learning methods are appropriate methods to improve the performance of classifiers for medical diagnosis. © Springer International Publishing Switzerland 2015.
Optimal power flow of two-terminal HVDC systems using backtracking search algorithm
(Elsevier Ltd, 2016) Ayan K.; Kiliç U.
This paper presents a solution technique for optimal power flow (OPF) of high-voltage direct current (HVDC) power systems using a backtracking search algorithm (BSA). BSA is a population-based evolutionary algorithm (EA), and it is not sensitive to initial conditions, contrary to most other meta-heuristic algorithms. The proposed algorithm is applied to three different test systems as follows: the modified 5-bus test system, the modified WSCC 9-bus test system, and the modified New England 39-bus test system. As a result of the simulations, minimum, maximum, and average production costs and CPU times are obtained for different cases of each of the three test systems. These results are also compared to those of the Artificial Bee Colony (ABC) algorithm, the Genetic Algorithm (GA), and the unified method provided in literature. In regard to the comparative results, it can be said that the proposed method has a shorter CPU time and is more efficient than the others. Thus, the applicability and efficiency of the proposed method in this field are demonstrated. © 2015 Elsevier Ltd. All rights reserved.
Exploring performance of instance selection methods in text sentiment classification
(Springer Verlag, 2016) Onan A.; Korukoğlu S.
Sentiment analysis is the process of extracting subjective information in source materials. Sentiment analysis is a subfield of web and text mining. One major problem encountered in these areas is overwhelming amount of data available. Hence, instance selection and feature selection become two essential tasks for achieving scalability in machine learning based sentiment classification. Instance selection is a data reduction technique which aims to eliminate redundant, noisy data from the training dataset so that training time can be reduced, scalability and generalization ability can be enhanced. This paper examines the predictive performance of fifteen benchmark instance selection methods for text classification domain. The instance selection methods are evaluated by decision tree classifier (C4.5 algorithm) and radial basis function networks in terms of classification accuracy and data reduction rates. The experimental results indicate that the highest classification accuracies on C4.5 algorithm are generally obtained by model class selection method, while the highest classification accuracies on radial basis function networks are obtained by nearest centroid neighbor edition. © Springer International Publishing Switzerland 2016.
A multiobjective weighted voting ensemble classifier based on differential evolution algorithm for text sentiment classification
(Elsevier Ltd, 2016) Onan A.; Korukoğlu S.; Bulut H.
Typically performed by supervised machine learning algorithms, sentiment analysis is highly useful for extracting subjective information from text documents online. Most approaches that use ensemble learning paradigms toward sentiment analysis involve feature engineering in order to enhance the predictive performance. In response, we sought to develop a paradigm of a multiobjective, optimization-based weighted voting scheme to assign appropriate weight values to classifiers and each output class based on the predictive performance of classification algorithms, all to enhance the predictive performance of sentiment classification. The proposed ensemble method is based on static classifier selection involving majority voting error and forward search, as well as a multiobjective differential evolution algorithm. Based on the static classifier selection scheme, our proposed ensemble method incorporates Bayesian logistic regression, naïve Bayes, linear discriminant analysis, logistic regression, and support vector machines as base learners, whose performance in terms of precision and recall values determines weight adjustment. Our experimental analysis of classification tasks, including sentiment analysis, software defect prediction, credit risk modeling, spam filtering, and semantic mapping, suggests that the proposed classification scheme can predict better than conventional ensemble learning methods such as AdaBoost, bagging, random subspace, and majority voting. Of all datasets examined, the laptop dataset showed the best classification accuracy (98.86%). © 2016 Elsevier Ltd
Ensemble of keyword extraction methods and classifiers in text classification
(Elsevier Ltd, 2016) Onan A.; Korukoǧlu S.; Bulut H.
Automatic keyword extraction is an important research direction in text mining, natural language processing and information retrieval. Keyword extraction enables us to represent text documents in a condensed way. The compact representation of documents can be helpful in several applications, such as automatic indexing, automatic summarization, automatic classification, clustering and filtering. For instance, text classification is a domain with high dimensional feature space challenge. Hence, extracting the most important/relevant words about the content of the document and using these keywords as the features can be extremely useful. In this regard, this study examines the predictive performance of five statistical keyword extraction methods (most frequent measure based keyword extraction, term frequency-inverse sentence frequency based keyword extraction, co-occurrence statistical information based keyword extraction, eccentricity-based keyword extraction and TextRank algorithm) on classification algorithms and ensemble methods for scientific text document classification (categorization). In the study, a comprehensive study of comparing base learning algorithms (Naïve Bayes, support vector machines, logistic regression and Random Forest) with five widely utilized ensemble methods (AdaBoost, Bagging, Dagging, Random Subspace and Majority Voting) is conducted. To the best of our knowledge, this is the first empirical analysis, which evaluates the effectiveness of statistical keyword extraction methods in conjunction with ensemble learning algorithms. The classification schemes are compared in terms of classification accuracy, F-measure and area under curve values. To validate the empirical analysis, two-way ANOVA test is employed. The experimental analysis indicates that Bagging ensemble of Random Forest with the most-frequent based keyword extraction method yields promising results for text classification. For ACM document collection, the highest average predictive performance (93.80%) is obtained with the utilization of the most frequent based keyword extraction method with Bagging ensemble of Random Forest algorithm. In general, Bagging and Random Subspace ensembles of Random Forest yield promising results. The empirical analysis indicates that the utilization of keyword-based representation of text documents in conjunction with ensemble learning can enhance the predictive performance and scalability of text classification schemes, which is of practical importance in the application fields of text classification. © 2016 Elsevier Ltd. All rights reserved.
Hybrid supervised clustering based ensemble scheme for text classification
(Emerald Group Publishing Ltd., 2017) Onan A.
Purpose: The immense quantity of available unstructured text documents serve as one of the largest source of information. Text classification can be an essential task for many purposes in information retrieval, such as document organization, text filtering and sentiment analysis. Ensemble learning has been extensively studied to construct efficient text classification schemes with higher predictive performance and generalization ability. The purpose of this paper is to provide diversity among the classification algorithms of ensemble, which is a key issue in the ensemble design. Design/methodology/approach: An ensemble scheme based on hybrid supervised clustering is presented for text classification. In the presented scheme, supervised hybrid clustering, which is based on cuckoo search algorithm and k-means, is introduced to partition the data samples of each class into clusters so that training subsets with higher diversities can be provided. Each classifier is trained on the diversified training subsets and the predictions of individual classifiers are combined by the majority voting rule. The predictive performance of the proposed classifier ensemble is compared to conventional classification algorithms (such as Naïve Bayes, logistic regression, support vector machines and C4.5 algorithm) and ensemble learning methods (such as AdaBoost, bagging and random subspace) using 11 text benchmarks. Findings: The experimental results indicate that the presented classifier ensemble outperforms the conventional classification algorithms and ensemble learning methods for text classification. Originality/value: The presented ensemble scheme is the first to use supervised clustering to obtain diverse ensemble for text classification © 2017, © Emerald Publishing Limited.
A feature selection model based on genetic rank aggregation for text sentiment classification
(SAGE Publications Ltd, 2017) Onan A.; KorukoGlu S.
Sentiment analysis is an important research direction of natural language processing, text mining and web mining which aims to extract subjective information in source materials. The main challenge encountered in machine learning method-based sentiment classification is the abundant amount of data available. This amount makes it difficult to train the learning algorithms in a feasible time and degrades the classification accuracy of the built model. Hence, feature selection becomes an essential task in developing robust and efficient classification models whilst reducing the training time. In text mining applications, individual filter-based feature selection methods have been widely utilized owing to their simplicity and relatively high performance. This paper presents an ensemble approach for feature selection, which aggregates the several individual feature lists obtained by the different feature selection methods so that a more robust and efficient feature subset can be obtained. In order to aggregate the individual feature lists, a genetic algorithm has been utilized. Experimental evaluations indicated that the proposed aggregation model is an efficient method and it outperforms individual filter-based feature selection methods on sentiment classification. © The Author(s) 2015.
Placement of Dg, Cb, and Tcsc in radial distribution system for power loss minimization using back-tracking search algorithm
(Springer Verlag, 2017) Fadel W.; Kilic U.; Taskin S.
The back-tracking search algorithm (BSA) is a new heuristic algorithm. BSA has two especially important properties: it is not sensitive to the initial value and has a single control parameter. This study presents the BSA-based optimal sizing and placement of distributed generations (DGs), capacitor banks (CBs), and thyristor-controlled series compensator (TCSC) in a radial distribution system (RDS). These elements are integrated separately and simultaneously in RDS. The objective function is power loss. The BSA is executed on IEEE 33 bus RDS. The obtained results are compared to a genetic algorithm (GA) and other algorithms in the literature. The results demonstrate that the BSA is more efficient and has the potential to find optimal solutions with less power loss. In this paper, optimal placement and sizing of DGs, TCSC, and CBs in a RDS is solved simultaneously using BSA for the first time. © 2016, Springer-Verlag Berlin Heidelberg.
Review spam detection based on psychological and linguistic features; [Psikolojik ve Dilbilimsel Özniteliklere Dayali Istenmeyen Inceleme Metni Belirleme]
(Institute of Electrical and Electronics Engineers Inc., 2018) Onan A.
With the advances in information and communication technologies, the immense quantity of review texts have become available on the Web. Review text can serve as an essential source of information for individual decision makers and business organizations. Some of the reviews shared on the Web may contain deceptive information to mislead the existing decision making process. In this study, we have presented a supervised learning based scheme for review spam detection. In the presented study, psychological and linguistic feature sets and their combinations are taken into consideration. In the study, the predictive performances of four conventional supervised learning methods (namely, Naive Bayes classifier, K-nearest neighbor algorithm, support vector machines and C4.5 algorithm) are evaluated on the different feature sets. © 2018 IEEE.
Investigating the effect of joint behavior on the optimum design of steel frames via hunting search algorithm
(Hong Kong Institute of Steel Construction, 2018) Doğan E.; Şeker S.; Polat Saka M.; Kozanoğlu C.
This study aims to carry out the effect of beam-to-column connections on the minimum weight design of steel plane frames. In the practical analysis of steel frames, end connections are assumed to be either fully restrained or pin-connected. However, experiments reveal that the real behavior is between these extremes and should be taken into account for the realistic design of structures. Hunting search algorithm is used for the automation of optimum design process. It is a numerical optimization method inspired by group hunting of animals such as wolves and lions. It is proven that it is a reliable and efficient technique for obtaining the solution of discrete structural optimization problems. Present design algorithm developed on the basis of hunting search algorithm selects w-sections for the members of semi rigid steel frame from the complete list of w-sections given in LRFD-AISC (Load and Resistance Factor Design, American Institute of Steel Construction). The design constraints are implemented from the specifications of the same code which covers serviceability and strength limitations. The selection of w-sections is carried out such that the design limitations are satisfied and the weight of semi-rigid frame is the minimum. In order to demonstrate its efficiency, three different steel frames are designed by the optimum design algorithm presented. The designs obtained by use of this algorithm are also compared with the ones produced by particle swarm optimization method. © 2018, Hong Kong Institute of Steel Construction. All rights reserved.
An ensemble scheme based on language function analysis and feature engineering for text genre classification
(SAGE Publications Ltd, 2018) Onan A.
Text genre classification is the process of identifying functional characteristics of text documents. The immense quantity of text documents available on the web can be properly filtered, organised and retrieved with the use of text genre classification, which may have potential use on several other tasks of natural language processing and information retrieval. Genre may refer to several aspects of text documents, such as function and purpose. The language function analysis (LFA) concentrates on single aspect of genres and it aims to classify text documents into three abstract classes, such as expressive, appellative and informative. Text genre classification is typically performed by supervised machine learning algorithms. The extraction of an efficient feature set to represent text documents is an essential task for building a robust classification scheme with high predictive performance. In addition, ensemble learning, which combines the outputs of individual classifiers to obtain a robust classification scheme, is a promising research field in machine learning research. In this regard, this article presents an extensive comparative analysis of different feature engineering schemes (such as features used in authorship attribution, linguistic features, character n-grams, part of speech n-grams and the frequency of the most discriminative words) and five different base learners (Naïve Bayes, support vector machines, logistic regression, k-nearest neighbour and Random Forest) in conjunction with ensemble learning methods (such as Boosting, Bagging and Random Subspace). Based on the empirical analysis, an ensemble classification scheme is presented, which integrates Random Subspace ensemble of Random Forest with four types of features (features used in authorship attribution, character n-grams, part of speech n-grams and the frequency of the most discriminative words). For LFA corpus, the highest average predictive performance obtained by the proposed scheme is 94.43%. © 2016, © The Author(s) 2016.

Browsing by Subject "Learning algorithms"

Results Per Page

Sort Options