Search results for: AI algorithm internal audit

Authors: Leehter Yao, Kuei-Song Weng, Cherng-Dir Huang

Abstract:

A fuzzy classifier using multiple ellipsoids approximating decision regions for classification is to be designed in this paper. An algorithm called Gustafson-Kessel algorithm (GKA) with an adaptive distance norm based on covariance matrices of prototype data points is adopted to learn the ellipsoids. GKA is able toadapt the distance norm to the underlying distribution of the prototypedata points except that the sizes of ellipsoids need to be determined a priori. To overcome GKA's inability to determine appropriate size ofellipsoid, the genetic algorithm (GA) is applied to learn the size ofellipsoid. With GA combined with GKA, it will be shown in this paper that the proposed method outperforms the benchmark algorithms as well as algorithms in the field.

Keywords: Ellipsoids, genetic algorithm, classification, fuzzyc-means (FCM)

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1688

Authors: Clark Van Dam, Gagan Mirchandani

Abstract:

The Continuously Adaptive Mean-Shift (CamShift) algorithm, incorporating scene depth information is combined with the l1-minimization sparse representation based method to form a hybrid kernel and state space-based tracking algorithm. We take advantage of the increased efficiency of the former with the robustness to occlusion property of the latter. A simple interchange scheme transfers control between algorithms based upon drift and occlusion likelihood. It is quantified by the projection of target candidates onto a depth map of the 2D scene obtained with a low cost stereo vision webcam. Results are improved tracking in terms of drift over each algorithm individually, in a challenging practical outdoor multiple occlusion test case.

Keywords: CamShift, l1-minimization, particle filter, stereo vision, video tracking.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2037

Authors: Mohamed H. S. Mohamed, Yang Xiao-zong, Liu Hong-wei, Wu Zhi-bo

Abstract:

The reliability of distributed systems and computer networks have been modeled by a probabilistic network or a graph G. Computing the residual connectedness reliability (RCR), denoted by R(G), under the node fault model is very useful, but is an NP-hard problem. Since it may need exponential time of the network size to compute the exact value of R(G), it is important to calculate its tight approximate value, especially its lower bound, at a moderate calculation time. In this paper, we propose an efficient algorithm for reliability lower bound of distributed systems with unreliable nodes. We also applied our algorithm to several typical classes of networks to evaluate the lower bounds and show the effectiveness of our algorithm.

Keywords: Distributed systems, probabilistic network, residual connectedness reliability, lower bound.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1673

Authors: Hanan Ahmed-Hosni Mahmoud, Nadia Al-Ghreimil

Abstract:

In-place sorting algorithms play an important role in many fields such as very large database systems, data warehouses, data mining, etc. Such algorithms maximize the size of data that can be processed in main memory without input/output operations. In this paper, a novel in-place sorting algorithm is presented. The algorithm comprises two phases; rearranging the input unsorted array in place, resulting segments that are ordered relative to each other but whose elements are yet to be sorted. The first phase requires linear time, while, in the second phase, elements of each segment are sorted inplace in the order of z log (z), where z is the size of the segment, and O(1) auxiliary storage. The algorithm performs, in the worst case, for an array of size n, an O(n log z) element comparisons and O(n log z) element moves. Further, no auxiliary arithmetic operations with indices are required. Besides these theoretical achievements of this algorithm, it is of practical interest, because of its simplicity. Experimental results also show that it outperforms other in-place sorting algorithms. Finally, the analysis of time and space complexity, and required number of moves are presented, along with the auxiliary storage requirements of the proposed algorithm.

Keywords: Auxiliary storage sorting, in-place sorting, sorting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1902

Authors: Anastasiia Yu. Timofeeva

Abstract:

Two new algorithms for nonparametric estimation of errors-in-variables models are proposed. The first algorithm is based on penalized regression spline. The spline is represented as a piecewise-linear function and for each linear portion orthogonal regression is estimated. This algorithm is iterative. The second algorithm involves locally weighted regression estimation. When the independent variable is measured with error such estimation is a complex nonlinear optimization problem. The simulation results have shown the advantage of the second algorithm under the assumption that true smoothing parameters values are known. Nevertheless the use of some indexes of fit to smoothing parameters selection gives the similar results and has an oversmoothing effect.

Keywords: Grade point average, orthogonal regression, penalized regression spline, locally weighted regression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2129

Authors: D.N. Zheng, J.X. Wang, Y.N. Zhao, Z.H. Yang

Abstract:

The Linear discriminant analysis (LDA) can be generalized into a nonlinear form - kernel LDA (KLDA) expediently by using the kernel functions. But KLDA is often referred to a general eigenvalue problem in singular case. To avoid this complication, this paper proposes an iterative algorithm for the two-class KLDA. The proposed KLDA is used as a nonlinear discriminant classifier, and the experiments show that it has a comparable performance with SVM.

Keywords: Linear discriminant analysis (LDA), kernel LDA (KLDA), conjugate gradient algorithm, nonlinear discriminant classifier.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1949

Authors: Ali Ridho Barakbah, Yasushi Kiyoki

Abstract:

This paper presents a new approach for image segmentation by applying Pillar-Kmeans algorithm. This segmentation process includes a new mechanism for clustering the elements of high-resolution images in order to improve precision and reduce computation time. The system applies K-means clustering to the image segmentation after optimized by Pillar Algorithm. The Pillar algorithm considers the pillars- placement which should be located as far as possible from each other to withstand against the pressure distribution of a roof, as identical to the number of centroids amongst the data distribution. This algorithm is able to optimize the K-means clustering for image segmentation in aspects of precision and computation time. It designates the initial centroids- positions by calculating the accumulated distance metric between each data point and all previous centroids, and then selects data points which have the maximum distance as new initial centroids. This algorithm distributes all initial centroids according to the maximum accumulated distance metric. This paper evaluates the proposed approach for image segmentation by comparing with K-means and Gaussian Mixture Model algorithm and involving RGB, HSV, HSL and CIELAB color spaces. The experimental results clarify the effectiveness of our approach to improve the segmentation quality in aspects of precision and computational time.

Keywords: Image segmentation, K-means clustering, Pillaralgorithm, color spaces.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3364

Authors: A.S. Rebaï, M. Elloumi

Abstract:

The Shortest Approximate Common Superstring (SACS) problem is : Given a set of strings f={w1, w2, ... , wn}, where no wi is an approximate substring of wj, i ≠ j, find a shortest string Sa, such that, every string of f is an approximate substring of Sa. When the number of the strings n>2, the SACS problem becomes NP-complete. In this paper, we present a greedy approximation SACS algorithm. Our algorithm is a 1/2-approximation for the SACS problem. It is of complexity O(n2*(l2+log(n))) in computing time, where n is the number of the strings and l is the length of a string. Our SACS algorithm is based on computation of the Length of the Approximate Longest Overlap (LALO).

Keywords: Shortest approximate common superstring, approximation algorithms, strings overlaps, complexities.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1499

Authors: Sheikh Muhammad Azam, Masood-ur-Rehman, Adnan Khalid Bhatti, Nadeem Daudpota

Abstract:

Very Large and/or computationally complex optimization problems sometimes require parallel or highperformance computing for achieving a reasonable time for computation. One of the most popular and most complicate problems of this family is “Traveling Salesman Problem". In this paper we have introduced a Branch & Bound based algorithm for the solution of such complicated problems. The main focus of the algorithm is to solve the “symmetric traveling salesman problem". We reviewed some of already available algorithms and felt that there is need of new algorithm which should give optimal solution or near to the optimal solution. On the basis of the use of logarithmic sampling, it was found that the proposed algorithm produced a relatively optimal solution for the problem and results excellent performance as compared with the traditional algorithms of this series.

Keywords: Parallel execution, symmetric traveling salesman problem, branch and bound algorithm, logarithmic sampling.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2329

Authors: Mahkameh S. Mostafavi

Abstract:

In this paper an algorithm is used to detect the color defects of ceramic tiles. First the image of a normal tile is clustered using GCMA; Genetic C-means Clustering Algorithm; those results in best cluster centers. C-means is a common clustering algorithm which optimizes an objective function, based on a measure between data points and the cluster centers in the data space. Here the objective function describes the mean square error. After finding the best centers, each pixel of the image is assigned to the cluster with closest cluster center. Then, the maximum errors of clusters are computed. For each cluster, max error is the maximum distance between its center and all the pixels which belong to it. After computing errors all the pixels of defected tile image are clustered based on the centers obtained from normal tile image in previous stage. Pixels which their distance from their cluster center is more than the maximum error of that cluster are considered as defected pixels.

Keywords: C-Means algorithm, color spaces, Genetic Algorithm, image clustering.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1640

Authors: Weng Ming Chu

Abstract:

When faced with stochastic networks with an uncertain duration for their activities, the securing of network completion time becomes problematical, not only because of the non-identical pdf of duration for each node, but also because of the interdependence of network paths. As evidenced by Adlakha & Kulkarni [1], many methods and algorithms have been put forward in attempt to resolve this issue, but most have encountered this same large-size network problem. Therefore, in this research, we focus on network reduction through a Series/Parallel combined mechanism. Our suggested algorithm, named the Activity Network Reduction Algorithm (ANRA), can efficiently transfer a large-size network into an S/P Irreducible Network (SPIN). SPIN can enhance stochastic network analysis, as well as serve as the judgment of symmetry for the Graph Theory.

Keywords: Series/Parallel network, Stochastic network, Network reduction, Interdictive Graph, Complexity Index.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1374

Authors: Poonam Garg, Aditya Shastri, D.C. Agarwal

Abstract:

With the exponential growth of networked system and application such as eCommerce, the demand for effective internet security is increasing. Cryptology is the science and study of systems for secret communication. It consists of two complementary fields of study: cryptography and cryptanalysis. The application of genetic algorithms in the cryptanalysis of knapsack ciphers is suggested by Spillman [7]. In order to improve the efficiency of genetic algorithm attack on knapsack cipher, the previously published attack was enhanced and re-implemented with variation of initial assumptions and results are compared with Spillman results. The experimental result of research indicates that the efficiency of genetic algorithm attack on knapsack cipher can be improved with variation of initial assumption.

Keywords: Genetic Algorithm, Knapsack cipher, Key search.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1684

Authors: Engy A. Mohamed, Yasser G. Hegazy

Abstract:

This paper presents a novel algorithm for modeling photovoltaic based distributed generators for the purpose of optimal planning of distribution networks. The proposed algorithm utilizes sequential Monte Carlo method in order to accurately consider the stochastic nature of photovoltaic based distributed generators. The proposed algorithm is implemented in MATLAB environment and the results obtained are presented and discussed.

Keywords: Comulative distribution function, distributed generation, Monte Carlo.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2478

Authors: Boukerch Haroun, Luo Qing Sheng, Li Hua Shi, Boukraa Sebti

Abstract:

In this paper we present the design and the implementation of a target tracking system where the target is set to be a moving person in a video sequence. The system can be applied easily as a vision system for mobile robot. The system is composed of two major parts the first is the detection of the person in the video frame using the SVM learning machine based on the “HOG” descriptors. The second part is the tracking of a moving person it’s done by using a combination of the Kalman filter and a modified version of the Camshift tracking algorithm by adding the target motion feature to the color feature, the experimental results had shown that the new algorithm had overcame the traditional Camshift algorithm in robustness and in case of occlusion.

Keywords: Camshift Algorithm, Computer Vision, Kalman Filter, Object tracking.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1325

Authors: E. Cenderelli, E. Bruno, G. Iacoviello, A. Lazzini

Abstract:

The financial sector is exposed to the risk of cyber-attacks like any other industrial sector. Furthermore, the topic of CybeRisk (cyber risk) has become particularly relevant given that Information Technology (IT) attacks have increased drastically in recent years, and cannot be stopped by single organizations requiring a response at international and national level. IT risk is never a matter purely for the IT manager, although he clearly plays a key role. A bank's risk management function requires a thorough understanding of the evolving risks as well as the tools and practical techniques available to address them. Upon the request of European and national legislation regarding CybeRisk in the financial system, banks are therefore called upon to strengthen the operational model for CybeRisk management. This will require an important change with a more intense collaboration with the structures that deal with information security for the development of an ad hoc system for the evaluation and control of this type of risk. The aim of the work is to propose a framework for the management and control of CybeRisk that will bridge the gap in the literature regarding the understanding and consideration of CybeRisk as an integral part of business management. The IT function has a strong relevance in the management of CybeRisk, which is perceived mainly as operational risk, but with a positive tendency on the part of risk management to the identification of CybeRisk assessment methods that are increasingly complete, quantitative and able to better describe the possible impacts on the business. The paper provides answers to the research questions: Is it possible to define a CybeRisk governance structure able to support the comparison between risk and security? How can the relationships between IT assets be integrated into a cyberisk assessment framework to guarantee a system of protection and risks control? From a methodological point of view, this research uses a case study approach. The choice of “Monte dei Paschi di Siena” was determined by the specific features of one of Italy’s biggest lenders. It is chosen to use an intensive research strategy: an in-depth study of reality. The case study methodology is an empirical approach to explore a complex and current phenomenon that develops over time. The use of cases has also the advantage of allowing the deepening of aspects concerning the "how" and "why" of contemporary events, on which the scholar has little control. The research bases on quantitative data and qualitative information obtained through semi-structured interviews of an open-ended nature and questionnaires to directors, members of the audit committee, risk, IT and compliance managers, and those responsible for internal audit function and anti-money laundering. The added value of the paper can be seen in the development of a framework based on a mapping of IT assets from which it is possible to identify their relationships for purposes of a more effective management and control of cyber risk.

Keywords: Bank, CybeRisk, information technology, risk management.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1415

Authors: Zhang Lei, Liu Gaotong

Abstract:

The atomization effect is an important factor of the heat transfer of liquid nitrogen spray. In this paper, two kinds of internal-mixing twin-fluid atomizers were design. According to the fracture theory and fluid mechanics, the model is established to simulate atomization effect. The results showed that: Internal-mixing atomizers, with the liquid nitrogen atomization size from 20um to 40um, have superior performance. Y-jet atomizer spray speed is greater than Multi-jet atomizer, and it can improve the efficiency of heat transfer between the liquid nitrogen and its spray object. Multi-jet atomizer atomization cone angle is about 30°, Y-jet atomizer atomization cone angle is about 20°. During atomizer selection, the size of the heat transfer area should be considered.

Keywords: Atomization, two-phase flow, atomizer, heat transfer.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2815

Authors: Dong-Hyung Kim, Ji Yeong Lee, Chang-Soo Han

Abstract:

This paper presents a sensor-based motion planning algorithm for 3-DOF car-like robots with a nonholonomic constraint. Similar to the classic Bug family algorithms, the proposed algorithm enables the car-like robot to navigate in a completely unknown environment using only the range sensor information. The car-like robot uses the local range sensor view to determine the local path so that it moves towards the goal. To guarantee that the robot can approach the goal, the two modes of motion are repeated, termed motion-to-goal and wall-following. The motion-to-goal behavior lets the robot directly move toward the goal, and the wall-following behavior makes the robot circumnavigate the obstacle boundary until it meets the leaving condition. For each behavior, the nonholonomic motion for the car-like robot is planned in terms of the instantaneous turning radius. The proposed algorithm is implemented to the real robot and the experimental results show the performance of proposed algorithm.

Keywords: Motion planning, car-like robot, bug algorithm, autonomous motion planning, nonholonomic constraint.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2227

Authors: Aji P. Wibawa, Andrew Nafalski, Neil Murray, Wayan F. Mahmudy

Abstract:

Since the one-to-one word translator does not have the facility to translate pragmatic aspects of Javanese, the parallel text alignment model described uses a phrase pair combination. The algorithm aligns the parallel text automatically from the beginning to the end of each sentence. Even though the results of the phrase pair combination outperform the previous algorithm, it is still inefficient. Recording all possible combinations consume more space in the database and time consuming. The original algorithm is modified by applying the edit distance coefficient to improve the data-storage efficiency. As a result, the data-storage consumption is 90% reduced as well as its learning period (42s).

Keywords: edit distance coefficient, Javanese, parallel text alignment, phrase pair combination

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1720

Authors: I. A. Farhat

Abstract:

The problem of optimal planning of multiple sources of distributed generation (DG) in distribution networks is treated in this paper using an improved Ant Colony Optimization algorithm (ACO). This objective of this problem is to determine the DG optimal size and location that in order to minimize the network real power losses. Considering the multiple sources of DG, both size and location are simultaneously optimized in a single run of the proposed ACO algorithm. The various practical constraints of the problem are taken into consideration by the problem formulation and the algorithm implementation. A radial power flow algorithm for distribution networks is adopted and applied to satisfy these constraints. To validate the proposed technique and demonstrate its effectiveness, the well-know 69-bus feeder standard test system is employed.cm.

Keywords: About Ant Colony Optimization (ACO), Distributed Generation (DG).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3270

Authors: Yongxin Yuan, Jiashang Jiang

Abstract:

In this paper the gradient based iterative algorithm is presented to solve the linear matrix equation AXB +CXTD = E, where X is unknown matrix, A,B,C,D,E are the given constant matrices. It is proved that if the equation has a solution, then the unique minimum norm solution can be obtained by choosing a special kind of initial matrices. Two numerical examples show that the introduced iterative algorithm is quite efficient.

Keywords: matrix equation, iterative algorithm, parameter estimation, minimum norm solution.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1547

Authors: Mohammad Ali Tabarzad, Caro Lucas, Ali Hamzeh

Abstract:

Adaptive Genetic Algorithms extend the Standard Gas to use dynamic procedures to apply evolutionary operators such as crossover, mutation and selection. In this paper, we try to propose a new adaptive genetic algorithm, which is based on the statistical information of the population as a guideline to tune its crossover, selection and mutation operators. This algorithms is called Statistical Genetic Algorithm and is compared with traditional GA in some benchmark problems.

Keywords: Genetic Algorithms, Statistical Information ofthe Population, PAUX, SSO.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1745

Authors: Reza Sirjani, Nobosse Tafem Bolan

Abstract:

Many optimization techniques available in the literature have been developed in order to solve the problem of voltage stability enhancement in power systems. However, there are a number of drawbacks in the use of previous techniques aimed at determining the optimal location and size of reactive compensators in a network. In this paper, an Improved Cuckoo Search algorithm is applied as an appropriate optimization algorithm to determine the optimum location and size of a Static Var Compensator (SVC) in a transmission network. The main objectives are voltage stability improvement and total cost minimization. The results of the presented technique are then compared with other available optimization techniques.

Keywords: Cuckoo search algorithm, optimization, power system, var compensators, voltage stability.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1342

Authors: S. Saehana, F. Iskandar, M. Abdullah, Khairurrijal

Abstract:

A numerical simulation of optimization all of electrospinning processing parameters to obtain smallest nanofiber diameter have been performed by employing genetic algorithm (GA). Fitness function in genetic algorithm methods, which was different for each parameter, was determined by simulation approach based on the Reneker’s model. Moreover, others genetic algorithm parameter, namely length of population, crossover and mutation were applied to get the optimum electrospinning processing parameters. In addition, minimum fiber diameter, 32 nm, was achieved from a simulation by applied the optimum parameters of electrospinning. This finding may be useful for process control and prediction of electrospun fiber production. In this paper, it is also compared between predicted parameters with some experimental results.

Keywords: Diameter, Electrospinning, GA, Nanofiber.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2950

Authors: N. Mpofu, M. Sears

Abstract:

In this paper, we propose a novel algorithm for delineating the endocardial wall from a human heart ultrasound scan. We assume that the gray levels in the ultrasound images are independent and identically distributed random variables with different Rician Inverse Gaussian (RiIG) distributions. Both synthetic and real clinical data will be used for testing the algorithm. Algorithm performance will be evaluated using the expert radiologist evaluation of a soft copy of an ultrasound scan during the scanning process and secondly, doctor’s conclusion after going through a printed copy of the same scan. Successful implementation of this algorithm should make it possible to differentiate normal from abnormal soft tissue and help disease identification, what stage the disease is in and how best to treat the patient. We hope that an automated system that uses this algorithm will be useful in public hospitals especially in Third World countries where problems such as shortage of skilled radiologists and shortage of ultrasound machines are common. These public hospitals are usually the first and last stop for most patients in these countries.

Keywords: Endorcardial Wall, Rician Inverse Distributions, Segmentation, Ultrasound Images.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1566

Authors: S. M. Kamruzzaman, Md. Monirul Islam

Abstract:

Although backpropagation ANNs generally predict better than decision trees do for pattern classification problems, they are often regarded as black boxes, i.e., their predictions cannot be explained as those of decision trees. In many applications, it is desirable to extract knowledge from trained ANNs for the users to gain a better understanding of how the networks solve the problems. A new rule extraction algorithm, called rule extraction from artificial neural networks (REANN) is proposed and implemented to extract symbolic rules from ANNs. A standard three-layer feedforward ANN is the basis of the algorithm. A four-phase training algorithm is proposed for backpropagation learning. Explicitness of the extracted rules is supported by comparing them to the symbolic rules generated by other methods. Extracted rules are comparable with other methods in terms of number of rules, average number of conditions for a rule, and predictive accuracy. Extensive experimental studies on several benchmarks classification problems, such as breast cancer, iris, diabetes, and season classification problems, demonstrate the effectiveness of the proposed approach with good generalization ability.

Keywords: Backpropagation, clustering algorithm, constructivealgorithm, continuous activation function, pruning algorithm, ruleextraction algorithm, symbolic rules.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1610

Authors: Fabio Fabris, Alex A. Freitas

Abstract:

Hierarchical classification is a special type of classification task where the class labels are organised into a hierarchy, with more generic class labels being ancestors of more specific ones. Meta-learning for classification-algorithm recommendation consists of recommending to the user a classification algorithm, from a pool of candidate algorithms, for a dataset, based on the past performance of the candidate algorithms in other datasets. Meta-learning is normally used in conventional, non-hierarchical classification. By contrast, this paper proposes a meta-learning approach for more challenging task of hierarchical classification, and evaluates it in a large number of bioinformatics datasets. Hierarchical classification is especially relevant for bioinformatics problems, as protein and gene functions tend to be organised into a hierarchy of class labels. This work proposes meta-learning approach for recommending the best hierarchical classification algorithm to a hierarchical classification dataset. This work’s contributions are: 1) proposing an algorithm for splitting hierarchical datasets into new datasets to increase the number of meta-instances, 2) proposing meta-features for hierarchical classification, and 3) interpreting decision-tree meta-models for hierarchical classification algorithm recommendation.

Keywords: Algorithm recommendation, meta-learning, bioinformatics, hierarchical classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1357

Authors: Christian Endisch, Christoph Hackl, Dierk Schröder

Abstract:

This paper presents an exact pruning algorithm with adaptive pruning interval for general dynamic neural networks (GDNN). GDNNs are artificial neural networks with internal dynamics. All layers have feedback connections with time delays to the same and to all other layers. The structure of the plant is unknown, so the identification process is started with a larger network architecture than necessary. During parameter optimization with the Levenberg- Marquardt (LM) algorithm irrelevant weights of the dynamic neural network are deleted in order to find a model for the plant as simple as possible. The weights to be pruned are found by direct evaluation of the training data within a sliding time window. The influence of pruning on the identification system depends on the network architecture at pruning time and the selected weight to be deleted. As the architecture of the model is changed drastically during the identification and pruning process, it is suggested to adapt the pruning interval online. Two system identification examples show the architecture selection ability of the proposed pruning approach.

Keywords: System identification, dynamic neural network, recurrentneural network, GDNN, optimization, Levenberg Marquardt, realtime recurrent learning, network pruning, quasi-online learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1934

Authors: Hao He, Shaoqian Li

Abstract:

In this paper, we study the cooperative communications where multiple cognitive radio (CR) transmit-receive pairs competitive maximize their own throughputs. In CR networks, the influences of primary users and the spectrum availability are usually different among CR users. Due to the existence of multiple relay nodes and the different spectrum availability, each CR transmit-receive pair should not only select the relay node but also choose the appropriate channel. For this distributed problem, we propose a game theoretic framework to formulate this problem and we apply a regret-matching learning algorithm which is leading to correlated equilibrium. We further formulate a modified regret-matching learning algorithm which is fully distributed and only use the local information of each CR transmit-receive pair. This modified algorithm is more practical and suitable for the cooperative communications in CR network. Simulation results show the algorithm convergence and the modified learning algorithm can achieve comparable performance to the original regretmatching learning algorithm.

Keywords: cognitive radio, cooperative communication, relay selection, channel choice, regret-matching learning, correlated equilibrium.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1666

Authors: M. T. Yazdani Sabouni, Rasaratnam Logendran

Abstract:

We consider a typical problem in the assembly of printed circuit boards (PCBs) in a two-machine flow shop system to simultaneously minimize the weighted sum of weighted tardiness and weighted flow time. The investigated problem is a group scheduling problem in which PCBs are assembled in groups and the interest is to find the best sequence of groups as well as the boards within each group to minimize the objective function value. The type of setup operation between any two board groups is characterized as carryover sequence-dependent setup time, which exactly matches with the real application of this problem. As a technical constraint, all of the boards must be kitted before the assembly operation starts (kitting operation) and by kitting staff. The main idea developed in this paper is to completely eliminate the role of kitting staff by assigning the task of kitting to the machine operator during the time he is idle which is referred to as integration of internal (machine) and external (kitting) setup times. Performing the kitting operation, which is a preparation process of the next set of boards while the other boards are currently being assembled, results in the boards to continuously enter the system or have dynamic arrival times. Consequently, a dynamic PCB assembly system is introduced for the first time in the assembly of PCBs, which also has characteristics similar to that of just-in-time manufacturing. The problem investigated is computationally very complex, meaning that finding the optimal solutions especially when the problem size gets larger is impossible. Thus, a heuristic based on Genetic Algorithm (GA) is employed. An example problem on the application of the GA developed is demonstrated and also numerical results of applying the GA on solving several instances are provided.

Keywords: Genetic algorithm, Dynamic PCB assembly, Carryover sequence-dependent setup times, Multi-objective.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1565

Authors: M. Kalamani, S. Valarmathy, M. Krishnamoorthi

Abstract:

In this paper, Least Mean Square (LMS) adaptive noise reduction algorithm is proposed to enhance the speech signal from the noisy speech. In this, the speech signal is enhanced by varying the step size as the function of the input signal. Objective and subjective measures are made under various noises for the proposed and existing algorithms. From the experimental results, it is seen that the proposed LMS adaptive noise reduction algorithm reduces Mean square Error (MSE) and Log Spectral Distance (LSD) as compared to that of the earlier methods under various noise conditions with different input SNR levels. In addition, the proposed algorithm increases the Peak Signal to Noise Ratio (PSNR) and Segmental SNR improvement (ΔSNRseg) values; improves the Mean Opinion Score (MOS) as compared to that of the various existing LMS adaptive noise reduction algorithms. From these experimental results, it is observed that the proposed LMS adaptive noise reduction algorithm reduces the speech distortion and residual noise as compared to that of the existing methods.

Keywords: LMS, speech enhancement, speech quality, residual noise.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2801