Статьи журнала - International Journal of Intelligent Systems and Applications
Все статьи: 1126
A swarm intelligence based chaotic morphological approach for software development cost estimation
Статья научная
In the last century, with the inception of various software development industries at around mid-1960’s, the complexities and size of the software have always been a major concern for the industries. The ad-hoc process of development has evolved into a standardized one due to the increase in the size and complexity of software projects. The standardized process of software development was further evolved to predict the overall cost required for the development before the software is actually built. To achieve the same, many cost estimation methodologies have already been successfully implemented, each with certain pros and cons. The present scenario demands even further refined and accurate predictions, which the above-said methods cease to provide. In this paper, we present a chaotically modified particle swarm optimization (CMPSO) based morphological learning approach to accurately estimate the cost incurred in the development process. The proposed approach focuses on a mathematical morphological (MM) framework based hybrid artificial neuron (also called dilation-erosion perceptron or DEP) with algebraic foundations in complete lattice theory (CLT). The proposed CMPSO-DEP model was tested on 5 well-known datasets of software projects with three popular performance metrics and the results were compared with the best existing models available in the literature.
Бесплатно
A trend analysis of machine learning research with topic models and Mann-Kendall test
Статья научная
This paper aims to systematically examine the literature of machine learning for the period of 1968~2017 to identify and analyze the research trends. A list of journals from well-established publishers ScienceDirect, Springer, JMLR, IEEE (approximately 23,365 journal articles) related to machine learning is used to prepare a content collection. To the best of our information, it is the first effort to comprehend the trend analysis in machine learning research with topic models: Latent Semantic Analysis (LSA), Latent Dirichlet Allocation (LDA), and LDA with Coherent Model (LDA_CM). The LDA_CM topic model gives the highest topic coherence amongst all topic models under consideration. This study provides a scientific ground that helps to overcome the subjectivity of collective opinion. The Mann-Kendall test is used to understand the trend of the topics. Our findings provide indicative of paradigmatic shifts in research methodology of significant patterns of topical prominence and the evolving research areas. It is used to highlight the evolution regarding the previous and recent trends in research topics in the area of machine learning. Understanding such an intellectual structure and future trends will assist the researchers to adopt the divergent developments of this research in one place. This paper analyzes the overall trends of the machine learning research since 1968, based on the latent topics identified in the period of 2007~2017 that may be helpful to the researchers exploring the recommended areas and publish their research articles.
Бесплатно
ABC Algorithm Based Interval Type-2 Fuzzy Logic Controller for an Inverted Pendulum
Статья научная
In this paper, a hybrid control technique is proposed for managing the variation of angle and velocity of the inverted pendulum. The proposed hybrid technique is the combination of ABC algorithm and interval type-2 Fuzzy Logic System (IT2FLS). The objective of the proposed hybrid control technique is to achieve the stability position of the pendulum. Here, the ABC algorithm is used to optimize the change of angle and change of velocity of the pendulum. With the optimized value, the optimal membership functions and the interference system are developed using IT2FLS. Using the ABC based IT2FLS, the position of the inverted pendulum is maintained towards the reference position. The proposed hybrid control technique is implemented in MATLAB/Simulink working platform and the control performances are evaluated.
Бесплатно
ANN model of border regions development: approach of closed systems
Статья научная
In this article we have suggested a new method of regional systems study that is based on model environmental influences isolation on their parameters dynamics. The presented model deepens greatly the investigation process of complex systems and allows defining clearly its functioning peculiarities without significant reduction in number of system characteristics as if we have simple technical or physical objects of knowledge. The described method, together with the statistical control data, is used for other social and economic objects research. The successful model testing in the form of artificial neural network model of Chernivtsi region static parameters has revealed the peculiarities of its interaction with European neighbors. In particular, for the first time we have defined their contribution to the increase of some social and economic indices on the period 2005-2015, that cannot be explained by other methods, such as correlation and regressive analysis. Applied use of the isolated investigation idea of the complex meso level systems together with the technology of data mining allows solving many actual tasks nominated by the regional administration practical workers.
Бесплатно
About Lyapunov exponents identification for systems with periodic coefficients
Статья научная
Lyapunov exponents (LE) identification prob-lem of dynamic systems with periodic coefficients is con-sidered under uncertainty. LE identification is based on the analysis of framework special class describing dy-namics of their change. Upper bound for the smallest LE and mobility limit for the large LE are obtained and the indicator set of the system is determined. The graphics criteria based on the analysis of framework special class features are proposed for an adequacy estimation of obtained LE estimations. The histogram method is applied to check for obtained estimation set. We show that the dynamic system can have the LE set.
Бесплатно
Accelerating Activation Function for 3-Satisfiability Logic Programming
Статья научная
This paper presents the technique for accelerating 3-Satisfiability (3-SAT) logic programming in Hopfield neural network. The core impetus for this work is to integrate activation function for doing 3-SAT logic programming in Hopfield neural network as a single hybrid network. In logic programming, the activation function can be used as a dynamic post optimization paradigm to transform the activation level of a unit (neuron) into an output signal. In this paper, we proposed Hyperbolic tangent activation function and Elliot symmetric activation function. Next, we compare the performance of proposed activation functions with a conventional function, namely McCulloch-Pitts function. In this study, we evaluate the performances between these functions through computer simulations. Microsoft Visual C++ 2013 was used as a platform for training, validating and testing of the network. We restrict our analysis to 3-Satisfiability (3-SAT) clauses. Moreover, evaluations are made between these activation functions to see the robustness via aspects of global solutions, global Hamming distance, and CPU time.
Бесплатно
Accelerating training of deep neural networks on GPU using CUDA
Статья научная
The development of fast and efficient training algorithms for Deep Neural Networks has been a subject of interest over the past few years because the biggest drawback of Deep Neural Networks is enormous cost in computation and large time is consumed to train the parameters of Deep Neural Networks. This aspect motivated several researchers to focus on recent advancements of hardware architectures and parallel programming models and paradigms for accelerating the training of Deep Neural Networks. We revisited the concepts and mechanisms of typical Deep Neural Network training algorithms such as Backpropagation Algorithm and Boltzmann Machine Algorithm and observed that the matrix multiplication constitutes major portion of the work-load for the Deep Neural Network training process because it is carried out for a huge number of times during the training of Deep Neural Networks. With the advent of many-core GPU technologies, a matrix multiplication can be done very efficiently in parallel and this helps a lot training a Deep Neural Network not consuming time as it used to be a few years ago. CUDA is one of the high performance parallel programming models to exploit the capabilities of modern many-core GPU systems. In this paper, we propose to modify Backpropagation Algorithm and Boltzmann Machine Algorithm with CUDA parallel matrix multiplication and test on many-core GPU system. Finally we discover that the planned strategies achieve very quick training of Deep Neural Networks than classic strategies.
Бесплатно
Acoustic Signal Classification from Monaural Recordings
Статья научная
Acoustic domain contains signals related to sound. Speech and music though are included in this domain, both the signals differ with various features. Features used for speech separation does not provide sufficient cue for music separation. This paper covers musical sound separation for monaural recordings. A system is proposed to classify singing voice and music from monaural recordings. For classification, time and frequency domain features along with Mel Frequency Cepstral Coefficients (MFCC) applied to input signal. Information carried by these signals permit to establish results Quantitative experimental results shows that the system performs the separation task successfully in monaural environment.
Бесплатно
Adaptation of Induced Fuzzy Cognitive Maps to the Problems Faced by the Power Loom Workers
Статья научная
The Indian textile industry has a significant presence in the economy as well as in the international textile economy. In this research Paper we study the socio economic problems faced by power loom workers in Avinashi in Tamilnadu, India, using Induced Fuzzy Cognitive Maps (IFCMs). We have interviewed 50 households in the study area using a linguistic questionnaire. As the problems faced by them at large, involved so much of feelings and uncertainties. We felt it to fit to use fuzzy theory in general and induced fuzzy cognitive maps in particular. For IFCMs is the best suited tool when the data is an unsupervised one.
Бесплатно
Статья научная
This research involved developing a surgical robot assistant using an articulated PUMA robot running on a linear or nonlinear axis. The research concentrated on studying the artificial intelligence based switching computed torque controller to localization of an endoscopic tool. Results show that the switching artificial nonlinear control algorithm is capable to design a stable controller. For this system, error was used as the performance metric. Positioning of the endoscopic manipulator relative to the world coordinate frame was possible to within 0.05 inch. Error in maintaining a constant point in space is evident during repositioning however this was caused by limitations in the robot arm.
Бесплатно
Adaptive Inverse Model of Nonlinear Systems
Статья научная
This paper proposes nonlinear adaptive filter-bank (NAFB) based algorithm for inverse modeling of nonlinear systems. Inverse modeling has been an important component for sensor linearization, adaptive control, channel equalization in communication system and active noise control. Under practical situations, the plant/system behaves nonlinearly which can be modeled as both parallel and cascaded structures of linear and nonlinear transfer functions. These linear and nonlinear transfer functions can be either static or dynamic, time variant or time invariant. The proposed NAFB algorithms are applied to generate the inverse model of different types of nonlinear systems and their convergence performances are evaluated. These nonlinear inverse models can be suitably applied to many engineering applications.
Бесплатно
Adaptive Neuro-Fuzzy Inferential Approach for the Diagnosis of Prostate Diseases
Статья научная
In this study, Adaptive Neuro-fuzzy Inferential System (ANFIS) is adapted for diagnosing prostate diseases. The system involves generating and tuning a fuzzy inference system to handle the imprecise terms used for describing prostate cases and severity. Several diagnostic variables were used to learn the feature statistics present in a typical data, while the trained model was validated and adapted for testing new prostate cases. A total of 335 data from patients’ records were collected at the Medi Moses Prostate Centre, Kumasi Ghana. The dataset was partitioned into 70% which was used for model training, and the other 30% was utilized in the validation phase. The proposed model was implemented in the MATLAB environment. Evaluation result from the proposed system demonstrated that the system achieved an accurate diagnostic result with an RMSE value of 11%. This indicates that the system has a relatively high accuracy and could be accepted for prostate diagnosis. Furthermore, the model was able to learn well and generalize the features in the data set, making the proposed ANFIS model suitable for new cases. Performance analysis showed that the ANFIS is well suited for handling the crispy values used in prostate diagnosis; thus, it can be extensively employed in other similar areas of medical diagnosis.
Бесплатно
Adaptive Observers for Linear Time-Varying Dynamic Objects with Uncertainty Estimation
Статья научная
The method for construction adaptive observ-ers (AO) time-varying linear dynamic objects at non-fulfillment of condition excitation constancy (EC) is pro-posed. Synthesis of the adaptive observer is given as the solution of two tasks. The solution first a problem is a choice of the constant matrix decreasing the effect of EC condition. Procedures for obtaining of this matrix are proposed. The matrix specifies restrictions for a vector of parameters AO. The solution of the second problem gives a method of design adaptive multiplicative algorithms in the presence of the obtained restrictions. Procedures for an estimation uncertainty in an object are proposed. They are based on obtaining of static models giving the forecast change of uncertainty. Optimum estimations of the uncertainty are obtained which minimize an error between outputs of the object and AO. An exponential dissipativity of adaptive system is proved. The results of the modeling confirming the effectiveness of designed methods and procedures are presented.
Бесплатно
Adaptive Observers with Uncertainty in Loop Tuning for Linear Time-Varying Dynamical Systems
Статья научная
The method of construction adaptive observers for linear time-varying dynamical systems with one input and an output is offered. Adaptive algorithms for identification are designed. Adaptive algorithms not realized as an adaptive system contains parametric uncertainty (PU). Realized adaptive algorithms of identification parameters system are offered. They on the procedure of the estimation PU and algorithm of signal adaptation are based. The algorithm of velocity change system parameters estimation is proposed. Estimations PU and its misalignments are obtained. Boundedness of trajectories an adaptive system is proved. Exponential stability conditions of the adaptive system are obtained. Iterative procedure of construction a parametric restrictions area is proposed. Simulation results have confirmed the efficiency of the method construction an adaptive observer.
Бесплатно
Статья научная
This paper proposes a novel metaheuristic optimization algorithm and suggests an adaptive artificial neural network controller that based on the proposed optimization algorithm. The purpose of the neural controller is to track desired proposed velocities and path trajectory with the minimum error, in the presence of mobile robot parameters time variation and system model uncertainties. The proposed controller consists of two sub-neural controllers; the kinematic neural feedback controller, and the dynamic neural feedback controller. The external feedback kinematic neural controller was responsible of generating the velocity tracking signals that track the mobile robot linear and angular velocities depending on the robot posture error, and the desired velocities. On the other hand, the internal dynamic neural controller has been used to enhance the mobile robot against parameters uncertainty, parameters time variation, and disturbance noise. However, the proposed grass root population-based metaheuristic optimization algorithm has been used to optimize the weights of the neural network to have the behavior of an adaptive nonlinear trajectory tracking controller of a differential drive wheeled mobile robot. The proposed controller shows a very good ability to prepare an appropriate dynamic control left and right torque signals to drive various mobile robot platforms using the same offline optimized weights. Grass root optimization algorithms have been used due to their unique characteristics especially, theirs derivative free, ability to optimize discretely and continuous nonlinear functions, and ability to escape of local minimum solutions.
Бесплатно
Статья научная
This paper presents a fault tolerant control (FTC) based on Radial Base Function Neural Network (RBFNN) using an adaptive control law for double star induction machine (DSIM) under broken rotor bars (BRB) fault in a squirrel-cage in order to improve its reliability and availability. The proposed FTC is designed to compensate for the default effect by maintaining acceptable performance in case of BRB. The sufficient condition for the stability of the closed-loop system in faulty operation is analyzed and verified using Lyapunov theory. To proof the performance and effectiveness of the proposed FTC, a comparative study within sliding mode control (SMC) is carried out. Obtained results show that the proposed FTC has a better robustness against the BRB fault.
Бесплатно
Adaptive Random Link PSO with Link Change Variations and Confinement Handling
Статья научная
Particle Swarm Optimization is swarm based optimization technique. Swarm consists of particles and the particles fly through the problem space in Particle Swarm Optimization (PSO). Confinement methods and parameters such as Inertia Weight, Neighborhood of the particle have major impact on PSO performance. The paper presents variations of the PSO with adaptive random link neighborhood. The work carried out considers linearly decreasing inertia weight and different confinement methods. The performance of adaptive random link PSO by geometrical updation of velocity with confinement methods is tested here.
Бесплатно
Adaptive algorithm design for cooperative hunting in multi-robots
Статья научная
The multi-robot cooperative planning is gained significant attention in recent past mainly for the evaders hunting task. In evaders hunting, the robot nodes required to recognize their other team members and considering their current positions and capabilities to catch the stationary or moving evaders effectively through the cooperating path planning approach. The primary challenge to design cooperative multi-robot evader hunting system is efficient and adaptive coordination of multiple autonomous mobile robots with less delay and communication overhead in presence of big-size obstacles. The current solutions suffered from repeated hunting problem under the inaccessible network conditions due to the presence of big-size obstacles and ineffective utilization of known nodes information. In this paper, to alleviate the problem of repeated hunting and inefficient catching of all evaders in the network, we proposed the adaptive Bio-inspired Neural Network (ABNN) using the new shunting equation with the capability of adaptive hunting of all evaders in the system. We design ABNN based on the implicit robot to predict the next path to catch evaders efficiently by real robots. The use of implicit robot helps to prevent the big sized evaders and efficiently utilize the evader’s information. The simulation results demonstrate that ABNN performs efficient evaders hunting under the presence of big size obstacles.
Бесплатно
Adaptive finite-time convergence fuzzy ARX-laguerre system estimation
Статья научная
Convergence speed for system identification and estimation is a popular topic for determining the kinematics and dynamic identification/estimation of the parameters of robot manipulators. In this paper, adaptive fuzzy inverse dynamic system estimation is used to improve robust modeling, especially for a serial links robot manipulator. The Lyapunov technique is used to analyze the convergence rate of the tracking error and increase the accuracy response of the parameter estimation. Performance of robot estimation is conducted, and the results show fast convergence of the proposed finite time technique for a 6-DOF robot manipulator.
Бесплатно
Статья научная
Due to freedom to express views, opinions, news, etc and easier method to disseminate the information to large population worldwide, social media platforms are inundated with big streaming data characterized by both short text and long normal text. Getting the glimpse of ongoing events happening over social media is quintessential from the viewpoint of understanding the trends, and for this, topic modeling is the most important step. With reference to increase in proliferation of big data streaming from social media platforms, it is crucial to perform large scale topic modeling to extract the topics dynamically in an online manner. This paper proposes an adaptive framework for dynamic topic modeling from big data using deep learning approach. Approach based on approximation of online latent semantic indexing constrained by regularization has been put forth. The model is designed using deep network of feed forward layers. The framework works in an adaptive manner in the sense that model is extracts incrementally according to streaming data and retrieves dynamic topics. In order to get the trends and evolution of topics, the framework supports temporal topic modeling, and enables to detect implicit and explicit aspects from sentences also.
Бесплатно
