Showing papers in "Computational Intelligence and Neuroscience in 2017"
TL;DR: The best model is the deep VGG16 model trained with transfer learning, which yields an overall accuracy of 90.4% on the hold-out test set.
Abstract: Automatic and accurate estimation of disease severity is essential for food security, disease management, and yield loss prediction. Deep learning, the latest breakthrough in computer vision, is promising for fine-grained disease severity classification, as the method avoids the labor-intensive feature engineering and threshold-based segmentation. Using the apple black rot images in the PlantVillage dataset, which are further annotated by botanists with four severity stages as ground truth, a series of deep convolutional neural networks are trained to diagnose the severity of the disease. The performances of shallow networks trained from scratch and deep models fine-tuned by transfer learning are evaluated systemically in this paper. The best model is the deep VGG16 model trained with transfer learning, which yields an overall accuracy of 90.4% on the hold-out test set. The proposed deep learning model may have great potential in disease control for modern agriculture.
412 citations
TL;DR: The first plant image dataset collected by mobile phone in natural scene is presented, which contains 10,000 images of 100 ornamental plant species in Beijing Forestry University campus, and a 26-layer deep learning model consisting of 8 residual building blocks is designed for large-scale plant classification in natural environment.
Abstract: Plant image identification has become an interdisciplinary focus in both botanical taxonomy and computer vision. The first plant image dataset collected by mobile phone in natural scene is presented, which contains 10,000 images of 100 ornamental plant species in Beijing Forestry University campus. A 26-layer deep learning model consisting of 8 residual building blocks is designed for large-scale plant classification in natural environment. The proposed model achieves a recognition rate of 91.78% on the BJFU100 dataset, demonstrating that deep learning is a promising technology for smart forestry.
197 citations
TL;DR: A novel ensemble method, called Bagging of Extrapolation Borderline-SMOTE SVM (BEBS), has been proposed in dealing with imbalanced data learning (IDL) problems, and is believed to be the first model combining ensemble of SVMs with borderline information for solving such condition.
Abstract: Class imbalance ubiquitously exists in real life, which has attracted much interest from various domains. Direct learning from imbalanced dataset may pose unsatisfying results overfocusing on the accuracy of identification and deriving a suboptimal model. Various methodologies have been developed in tackling this problem including sampling, cost-sensitive, and other hybrid ones. However, the samples near the decision boundary which contain more discriminative information should be valued and the skew of the boundary would be corrected by constructing synthetic samples. Inspired by the truth and sense of geometry, we designed a new synthetic minority oversampling technique to incorporate the borderline information. What is more, ensemble model always tends to capture more complicated and robust decision boundary in practice. Taking these factors into considerations, a novel ensemble method, called Bagging of Extrapolation Borderline-SMOTE SVM (BEBS), has been proposed in dealing with imbalanced data learning (IDL) problems. Experiments on open access datasets showed significant superior performance using our model and a persuasive and intuitive explanation behind the method was illustrated. As far as we know, this is the first model combining ensemble of SVMs with borderline information for solving such condition.
143 citations
TL;DR: It is shown that the combination between advanced deep learning and novel signature can be a robust solution to overcome NILM's issues and improve the performance of load identification.
Abstract: Monitoring electricity consumption in the home is an important way to help reduce energy usage. Nonintrusive Load Monitoring (NILM) is existing technique which helps us monitor electricity consumption effectively and costly. NILM is a promising approach to obtain estimates of the electrical power consumption of individual appliances from aggregate measurements of voltage and/or current in the distribution system. Among the previous studies, Hidden Markov Model (HMM) based models have been studied very much. However, increasing appliances, multistate of appliances, and similar power consumption of appliances are three big issues in NILM recently. In this paper, we address these problems through providing our contributions as follows. First, we proposed state-of-the-art energy disaggregation based on Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) model and additional advanced deep learning. Second, we proposed a novel signature to improve classification performance of the proposed model in multistate appliance case. We applied the proposed model on two datasets such as UK-DALE and REDD. Via our experimental results, we have confirmed that our model outperforms the advanced model. Thus, we show that our combination between advanced deep learning and novel signature can be a robust solution to overcome NILM’s issues and improve the performance of load identification.
117 citations
TL;DR: A novel active semisupervised CNN algorithm is presented that uses the active learning to query the most informative and reliable samples in the unlabelled samples to extend the initial training dataset and adds a new regularization term into the loss function of CNN.
Abstract: Convolutional neural network (CNN) can be applied in synthetic aperture radar (SAR) object recognition for achieving good performance. However, it requires a large number of the labelled samples in its training phase, and therefore its performance could decrease dramatically when the labelled samples are insufficient. To solve this problem, in this paper, we present a novel active semisupervised CNN algorithm. First, the active learning is used to query the most informative and reliable samples in the unlabelled samples to extend the initial training dataset. Next, a semisupervised method is developed by adding a new regularization term into the loss function of CNN. As a result, the class probability information contained in the unlabelled samples can be maximally utilized. The experimental results on the MSTAR database demonstrate the effectiveness of the proposed algorithm despite the lack of the initial labelled samples.
117 citations
TL;DR: This article proposes a new crossover operator for traveling salesman problem to minimize the total distance and is linked with path representation, which is the most natural way to represent a legal tour.
Abstract: Genetic algorithms are evolutionary techniques used for optimization purposes according to survival of the fittest idea. These methods do not ensure optimal solutions; however, they give good approximation usually in time. The genetic algorithms are useful for NP-hard problems, especially the traveling salesman problem. The genetic algorithm depends on selection criteria, crossover, and mutation operators. To tackle the traveling salesman problem using genetic algorithms, there are various representations such as binary, path, adjacency, ordinal, and matrix representations. In this article, we propose a new crossover operator for traveling salesman problem to minimize the total distance. This approach has been linked with path representation, which is the most natural way to represent a legal tour. Computational results are also reported with some traditional path representation methods like partially mapped and order crossovers along with new cycle crossover operator for some benchmark TSPLIB instances and found improvements.
105 citations
TL;DR: A hybrid semi-supervised anomaly detection model for high-dimensional data that consists of a deep autoencoder (DAE) and an ensemble k-nearest neighbor graphs- (K-NNG-) based anomaly detector is proposed.
Abstract: Anomaly detection, which aims to identify observations that deviate from a nominal sample, is a challenging task for high-dimensional data. Traditional distance-based anomaly detection methods compute the neighborhood distance between each observation and suffer from the curse of dimensionality in high-dimensional space; for example, the distances between any pair of samples are similar and each sample may perform like an outlier. In this paper, we propose a hybrid semi-supervised anomaly detection model for high-dimensional data that consists of two parts: a deep autoencoder (DAE) and an ensemble -nearest neighbor graphs- (-NNG-) based anomaly detector. Benefiting from the ability of nonlinear mapping, the DAE is first trained to learn the intrinsic features of a high-dimensional dataset to represent the high-dimensional data in a more compact subspace. Several nonparametric KNN-based anomaly detectors are then built from different subsets that are randomly sampled from the whole dataset. The final prediction is made by all the anomaly detectors. The performance of the proposed method is evaluated on several real-life datasets, and the results confirm that the proposed hybrid model improves the detection accuracy and reduces the computational complexity.
102 citations
TL;DR: Left and right lower limb activation patterns were found to be highly similar during both imagery and execution, indicating that higher resolution imaging, advanced signal processing, or improved subject training may be required to reliably distinguish them.
Abstract: Motor-activity-related mental tasks are widely adopted for brain-computer interfaces (BCIs) as they are a natural extension of movement intention, requiring no training to evoke brain activity. The ideal BCI aims to eliminate neuromuscular movement, making motor imagery tasks, or imagined actions with no muscle movement, good candidates. This study explores cortical activation differences between motor imagery and motor execution for both upper and lower limbs using functional near-infrared spectroscopy (fNIRS). Four simple finger- or toe-tapping tasks (left hand, right hand, left foot, and right foot) were performed with both motor imagery and motor execution and compared to resting state. Significant activation was found during all four motor imagery tasks, indicating that they can be detected via fNIRS. Motor execution produced higher activation levels, a faster response, and a different spatial distribution compared to motor imagery, which should be taken into account when designing an imagery-based BCI. When comparing left versus right, upper limb tasks are the most clearly distinguishable, particularly during motor execution. Left and right lower limb activation patterns were found to be highly similar during both imagery and execution, indicating that higher resolution imaging, advanced signal processing, or improved subject training may be required to reliably distinguish them.
98 citations
TL;DR: A patient-specific epileptic seizure predication method relying on the common spatial pattern- (CSP-) based feature extraction of scalp electroencephalogram (sEEG) signals to train a linear discriminant analysis classifier, which is then employed in the testing phase.
Abstract: This paper presents a patient-specific epileptic seizure predication method relying on the common spatial pattern- (CSP-) based feature extraction of scalp electroencephalogram (sEEG) signals. Multichannel EEG signals are traced and segmented into overlapping segments for both preictal and interictal intervals. The features extracted using CSP are used for training a linear discriminant analysis classifier, which is then employed in the testing phase. A leave-one-out cross-validation strategy is adopted in the experiments. The experimental results for seizure prediction obtained from the records of 24 patients from the CHB-MIT database reveal that the proposed predictor can achieve an average sensitivity of 0.89, an average false prediction rate of 0.39, and an average prediction time of 68.71 minutes using a 120-minute prediction horizon.
97 citations
TL;DR: In this work a grey wolf optimizer is proposed for the design of modular granular neural networks, and the results are compared against a genetic algorithm and a firefly algorithm in order to know which of these techniques provides better results when applied to human recognition.
Abstract: A grey wolf optimizer for modular neural network (MNN) with a granular approach is proposed. The proposed method performs optimal granulation of data and design of modular neural networks architectures to perform human recognition, and to prove its effectiveness benchmark databases of ear, iris, and face biometric measures are used to perform tests and comparisons against other works. The design of a modular granular neural network (MGNN) consists in finding optimal parameters of its architecture; these parameters are the number of subgranules, percentage of data for the training phase, learning algorithm, goal error, number of hidden layers, and their number of neurons. Nowadays, there is a great variety of approaches and new techniques within the evolutionary computing area, and these approaches and techniques have emerged to help find optimal solutions to problems or models and bioinspired algorithms are part of this area. In this work a grey wolf optimizer is proposed for the design of modular granular neural networks, and the results are compared against a genetic algorithm and a firefly algorithm in order to know which of these techniques provides better results when applied to human recognition.
83 citations
TL;DR: How RNN-based autoencoders outperform statistical approaches up to an absolute improvement of 16.4% average F-measure over the three databases is shown.
Abstract: In the emerging field of acoustic novelty detection, most research efforts are devoted to probabilistic approaches such as mixture models or state-space models. Only recent studies introduced (pseudo-)generative models for acoustic novelty detection with recurrent neural networks in the form of an autoencoder. In these approaches, auditory spectral features of the next short term frame are predicted from the previous frames by means of Long-Short Term Memory recurrent denoising autoencoders. The reconstruction error between the input and the output of the autoencoder is used as activation signal to detect novel events. There is no evidence of studies focused on comparing previous efforts to automatically recognize novel events from audio signals and giving a broad and in depth evaluation of recurrent neural network-based autoencoders. The present contribution aims to consistently evaluate our recent novel approaches to fill this white spot in the literature and provide insight by extensive evaluations carried out on three databases: A3Novelty, PASCAL CHiME, and PROMETHEUS. Besides providing an extensive analysis of novel and state-of-the-art methods, the article shows how RNN-based autoencoders outperform statistical approaches up to an absolute improvement of 16.4% average F-measure over the three databases.
TL;DR: Two multimodal fusion methods between brain and peripheral signals for emotion recognition based on two decision-level fusion methods of both EEG and facial expression detections by using a sum rule or a production rule are proposed.
Abstract: This paper proposes two multimodal fusion methods between brain and peripheral signals for emotion recognition. The input signals are electroencephalogram and facial expression. The stimuli are based on a subset of movie clips that correspond to four specific areas of valance-arousal emotional space (happiness, neutral, sadness, and fear). For facial expression detection, four basic emotion states (happiness, neutral, sadness, and fear) are detected by a neural network classifier. For EEG detection, four basic emotion states and three emotion intensity levels (strong, ordinary, and weak) are detected by two support vector machines (SVM) classifiers, respectively. Emotion recognition is based on two decision-level fusion methods of both EEG and facial expression detections by using a sum rule or a production rule. Twenty healthy subjects attended two experiments. The results show that the accuracies of two multimodal fusion detections are 81.25% and 82.75%, respectively, which are both higher than that of facial expression (74.38%) or EEG detection (66.88%). The combination of facial expressions and EEG information for emotion recognition compensates for their defects as single information sources.
TL;DR: An ensemble of random deep belief networks (RDBN) method for speech emotion recognition that firstly extracts the low level features of the input speech signal and then applies them to construct lots of random subspaces.
Abstract: Now the human emotions can be recognized from speech signals using machine learning methods; however, they are challenged by the lower recognition accuracies in real applications due to lack of the rich representation ability. Deep belief networks (DBN) can automatically discover the multiple levels of representations in speech signals. To make full of its advantages, this paper presents an ensemble of random deep belief networks (RDBN) method for speech emotion recognition. It firstly extracts the low level features of the input speech signal and then applies them to construct lots of random subspaces. Each random subspace is then provided for DBN to yield the higher level features as the input of the classifier to output an emotion label. All outputted emotion labels are then fused through the majority voting to decide the final emotion label for the input speech signal. The conducted experimental results on benchmark speech emotion databases show that RDBN has better accuracy than the compared methods for speech emotion recognition.
TL;DR: The modified discrete grey wolf optimizer algorithm (MDGWO), which improves on the optimal solution updating mechanism of the search agent by the weights, is put forward, which has advantages over the latter four in terms of image segmentation quality and objective function values and their stability.
Abstract: The computation of image segmentation has become more complicated with the increasing number of thresholds, and the option and application of the thresholds in image thresholding fields have become an NP problem at the same time. The paper puts forward the modified discrete grey wolf optimizer algorithm (MDGWO), which improves on the optimal solution updating mechanism of the search agent by the weights. Taking Kapurźs entropy as the optimized function and based on the discreteness of threshold in image segmentation, the paper firstly discretizes the grey wolf optimizer (GWO) and then proposes a new attack strategy by using the weight coefficient to replace the search formula for optimal solution used in the original algorithm. The experimental results show that MDGWO can search out the optimal thresholds efficiently and precisely, which are very close to the result examined by exhaustive searches. In comparison with the electromagnetism optimization (EMO), the differential evolution (DE), the Artifical Bee Colony (ABC), and the classical GWO, it is concluded that MDGWO has advantages over the latter four in terms of image segmentation quality and objective function values and their stability.
TL;DR: This article reviews the major techniques needed for developing BRI systems, and describes a few BRI applications, including wheelchairs, manipulators, drones, and humanoid robots with respect to synchronous and asynchronous BCI-based techniques.
Abstract: The most popular noninvasive Brain Robot Interaction (BRI) technology uses the electroencephalogram- (EEG-) based Brain Computer Interface (BCI), to serve as an additional communication channel, for robot control via brainwaves. This technology is promising for elderly or disabled patient assistance with daily life. The key issue of a BRI system is to identify human mental activities, by decoding brainwaves, acquired with an EEG device. Compared with other BCI applications, such as word speller, the development of these applications may be more challenging since control of robot systems via brainwaves must consider surrounding environment feedback in real-time, robot mechanical kinematics, and dynamics, as well as robot control architecture and behavior. This article reviews the major techniques needed for developing BRI systems. In this review article, we first briefly introduce the background and development of mind-controlled robot technologies. Second, we discuss the EEG-based brain signal models with respect to generating principles, evoking mechanisms, and experimental paradigms. Subsequently, we review in detail commonly used methods for decoding brain signals, namely, preprocessing, feature extraction, and feature classification, and summarize several typical application examples. Next, we describe a few BRI applications, including wheelchairs, manipulators, drones, and humanoid robots with respect to synchronous and asynchronous BCI-based techniques. Finally, we address some existing problems and challenges with future BRI techniques.
TL;DR: An improved dynamic BINN is proposed and a virtual target is proposed in the path planning method to ensure that the AUV can move to the real target effectively and avoid big-size obstacles automatically.
Abstract: Real-time path planning for autonomous underwater vehicle (AUV) is a very difficult and challenging task. Bioinspired neural network (BINN) has been used to deal with this problem for its many distinct advantages: that is, no learning process is needed and realization is also easy. However, there are some shortcomings when BINN is applied to AUV path planning in a three-dimensional (3D) unknown environment, including complex computing problem when the environment is very large and repeated path problem when the size of obstacles is bigger than the detection range of sensors. To deal with these problems, an improved dynamic BINN is proposed in this paper. In this proposed method, the AUV is regarded as the core of the BINN and the size of the BINN is based on the detection range of sensors. Then the BINN will move with the AUV and the computing could be reduced. A virtual target is proposed in the path planning method to ensure that the AUV can move to the real target effectively and avoid big-size obstacles automatically. Furthermore, a target attractor concept is introduced to improve the computing efficiency of neural activities. Finally, some experiments are conducted under various 3D underwater environments. The experimental results show that the proposed BINN based method can deal with the real-time path planning problem for AUV efficiently.
TL;DR: The results of the experiment and security analysis show that the algorithm not only has a large key space and strong sensitivity to the key but can also effectively resist attack operations such as statistical analysis and exhaustive analysis.
Abstract: With the help of the fact that chaos is sensitive to initial conditions and pseudorandomness, combined with the spatial configurations in the DNA molecule's inherent and unique information processing ability, a novel image encryption algorithm based on bit permutation and dynamic DNA encoding is proposed here. The algorithm first uses Keccak to calculate the hash value for a given DNA sequence as the initial value of a chaotic map; second, it uses a chaotic sequence to scramble the image pixel locations, and the butterfly network is used to implement the bit permutation. Then, the image is coded into a DNA matrix dynamic, and an algebraic operation is performed with the DNA sequence to realize the substitution of the pixels, which further improves the security of the encryption. Finally, the confusion and diffusion properties of the algorithm are further enhanced by the operation of the DNA sequence and the ciphertext feedback. The results of the experiment and security analysis show that the algorithm not only has a large key space and strong sensitivity to the key but can also effectively resist attack operations such as statistical analysis and exhaustive analysis.
TL;DR: In this paper, in order to classify electrical faults in radial distribution systems, a particle swarm optimization (PSO) based support vector machine (SVM) classifier has been proposed, able to select appropriate input features and optimize SVM parameters to increase classification accuracy.
Abstract: Fast and accurate fault classification is essential to power system operations. In this paper, in order to classify electrical faults in radial distribution systems, a particle swarm optimization (PSO) based support vector machine (SVM) classifier has been proposed. The proposed PSO based SVM classifier is able to select appropriate input features and optimize SVM parameters to increase classification accuracy. Further, a time-domain reflectometry (TDR) method with a pseudorandom binary sequence (PRBS) stimulus has been used to generate a dataset for purposes of classification. The proposed technique has been tested on a typical radial distribution network to identify ten different types of faults considering 12 given input features generated by using Simulink software and MATLAB Toolbox. The success rate of the SVM classifier is over 97%, which demonstrates the effectiveness and high efficiency of the developed method.
TL;DR: A novel eye control system with integrating both mouse and keyboard functions is proposed in this paper, which focuses on providing a simple and convenient interactive mode by only using user's eye.
Abstract: Eye movement can be regarded as a pivotal real-time input medium for human-computer communication, which is especially important for people with physical disability. In order to improve the reliability, mobility, and usability of eye tracking technique in user-computer dialogue, a novel eye control system with integrating both mouse and keyboard functions is proposed in this paper. The proposed system focuses on providing a simple and convenient interactive mode by only using user's eye. The usage flow of the proposed system is designed to perfectly follow human natural habits. Additionally, a magnifier module is proposed to allow the accurate operation. In the experiment, two interactive tasks with different difficulty (searching article and browsing multimedia web) were done to compare the proposed eye control tool with an existing system. The Technology Acceptance Model (TAM) measures are used to evaluate the perceived effectiveness of our system. It is demonstrated that the proposed system is very effective with regard to usability and interface design.
TL;DR: This paper explores the applicability of deep learning techniques for detecting deviations from the norm in behavioral patterns of vessels (outliers) as they are tracked from an OTH radar using the nonlinear mapping capabilities of deep stacked autoencoders in combination with density-based clustering.
Abstract: Detection of outliers in radar signals is a considerable challenge in maritime surveillance applications. High-Frequency Surface-Wave (HFSW) radars have attracted significant interest as potential tools for long-range target identification and outlier detection at over-the-horizon (OTH) distances. However, a number of disadvantages, such as their low spatial resolution and presence of clutter, have a negative impact on their accuracy. In this paper, we explore the applicability of deep learning techniques for detecting deviations from the norm in behavioral patterns of vessels (outliers) as they are tracked from an OTH radar. The proposed methodology exploits the nonlinear mapping capabilities of deep stacked autoencoders in combination with density-based clustering. A comparative experimental evaluation of the approach shows promising results in terms of the proposed methodology's performance.
TL;DR: A robust and adaptive online gradient learning method for long short-term memory (LSTM) to predict time series with outliers, which tunes the learning rate of the stochastic gradient algorithm adaptively in the process of prediction, which reduces the adverse effect of outliers.
Abstract: Online time series prediction is the mainstream method in a wide range of fields, ranging from speech analysis and noise cancelation to stock market analysis. However, the data often contains many outliers with the increasing length of time series in real world. These outliers can mislead the learned model if treated as normal points in the process of prediction. To address this issue, in this paper, we propose a robust and adaptive online gradient learning method, RoAdam (Robust Adam), for long short-term memory (LSTM) to predict time series with outliers. This method tunes the learning rate of the stochastic gradient algorithm adaptively in the process of prediction, which reduces the adverse effect of outliers. It tracks the relative prediction error of the loss function with a weighted average through modifying Adam, a popular stochastic gradient method algorithm for training deep neural networks. In our algorithm, the large value of the relative prediction error corresponds to a small learning rate, and vice versa. The experiments on both synthetic data and real time series show that our method achieves better performance compared to the existing methods based on LSTM.
TL;DR: A bee evolutionary guiding nondominated sorting genetic algorithm II (BEG-NSGA-II) for multiobjective FJSP (MO-FJSP) with the objectives to minimize the maximal completion time, the workload of the most loaded machine, and the total workload of all machines.
Abstract: Flexible job-shop scheduling problem (FJSP) is an NP-hard puzzle which inherits the job-shop scheduling problem (JSP) characteristics. This paper presents a bee evolutionary guiding nondominated sorting genetic algorithm II (BEG-NSGA-II) for multiobjective FJSP (MO-FJSP) with the objectives to minimize the maximal completion time, the workload of the most loaded machine, and the total workload of all machines. It adopts a two-stage optimization mechanism during the optimizing process. In the first stage, the NSGA-II algorithm with T iteration times is first used to obtain the initial population N, in which a bee evolutionary guiding scheme is presented to exploit the solution space extensively. In the second stage, the NSGA-II algorithm with GEN iteration times is used again to obtain the Pareto-optimal solutions. In order to enhance the searching ability and avoid the premature convergence, an updating mechanism is employed in this stage. More specifically, its population consists of three parts, and each of them changes with the iteration times. What is more, numerical simulations are carried out which are based on some published benchmark instances. Finally, the effectiveness of the proposed BEG-NSGA-II algorithm is shown by comparing the experimental results and the results of some well-known algorithms already existed.
TL;DR: A novel image enhancement method via Mean and Variance based Subimage Histogram Equalization (MVSIHE), which effectively increases the contrast of the input image with brightness and details well preserved compared with some other methods based on histogram equalization (HE).
Abstract: This paper puts forward a novel image enhancement method via Mean and Variance based Subimage Histogram Equalization (MVSIHE), which effectively increases the contrast of the input image with brightness and details well preserved compared with some other methods based on histogram equalization (HE). Firstly, the histogram of input image is divided into four segments based on the mean and variance of luminance component, and the histogram bins of each segment are modified and equalized, respectively. Secondly, the result is obtained via the concatenation of the processed subhistograms. Lastly, the normalization method is deployed on intensity levels, and the integration of the processed image with the input image is performed. 100 benchmark images from a public image database named CVG-UGR-Database are used for comparison with other state-of-the-art methods. The experiment results show that the algorithm can not only enhance image information effectively but also well preserve brightness and details of the original image.
TL;DR: The system takes real-time utilization of resources and feeds utilization values into several buffers based on the type of Resources and time span size and selects a model based on minimum Akaike Information Criterion values.
Abstract: Infrastructure as a Service (IaaS) cloud provides resources as a service from a pool of compute, network, and storage resources. Cloud providers can manage their resource usage by knowing future usage demand from the current and past usage patterns of resources. Resource usage prediction is of great importance for dynamic scaling of cloud resources to achieve efficiency in terms of cost and energy consumption while keeping quality of service. The purpose of this paper is to present a real-time resource usage prediction system. The system takes real-time utilization of resources and feeds utilization values into several buffers based on the type of resources and time span size. Buffers are read by R language based statistical system. These buffers’ data are checked to determine whether their data follows Gaussian distribution or not. In case of following Gaussian distribution, Autoregressive Integrated Moving Average (ARIMA) is applied; otherwise Autoregressive Neural Network (AR-NN) is applied. In ARIMA process, a model is selected based on minimum Akaike Information Criterion (AIC) values. Similarly, in AR-NN process, a network with the lowest Network Information Criterion (NIC) value is selected. We have evaluated our system with real traces of CPU utilization of an IaaS cloud of one hundred and twenty servers.
TL;DR: Digital image encryption employs a process of transforming the image pixel gray value by using chaotic sequence scrambling image pixel location and establishing superchaotic mapping, which maps quaternary sequences and DNA sequences, and by combining with the logic of the transformation between DNA sequences.
Abstract: Image encryption technology is one of the main means to ensure the safety of image information. Using the characteristics of chaos, such as randomness, regularity, ergodicity, and initial value sensitiveness, combined with the unique space conformation of DNA molecules and their unique information storage and processing ability, an efficient method for image encryption based on the chaos theory and a DNA sequence database is proposed. In this paper, digital image encryption employs a process of transforming the image pixel gray value by using chaotic sequence scrambling image pixel location and establishing superchaotic mapping, which maps quaternary sequences and DNA sequences, and by combining with the logic of the transformation between DNA sequences. The bases are replaced under the displaced rules by using DNA coding in a certain number of iterations that are based on the enhanced quaternary hyperchaotic sequence; the sequence is generated by Chen chaos. The cipher feedback mode and chaos iteration are employed in the encryption process to enhance the confusion and diffusion properties of the algorithm. Theoretical analysis and experimental results show that the proposed scheme not only demonstrates excellent encryption but also effectively resists chosen-plaintext attack, statistical attack, and differential attack.
TL;DR: In this paper, a new novel swarm intelligence algorithm, namely, dandelion algorithm (DA), is proposed for global optimization of complex functions, and simulations show that the proposed algorithm seems much superior to other algorithms.
Abstract: Inspired by the behavior of dandelion sowing, a new novel swarm intelligence algorithm, namely, dandelion algorithm (DA), is proposed for global optimization of complex functions in this paper. In DA, the dandelion population will be divided into two subpopulations, and different subpopulations will undergo different sowing behaviors. Moreover, another sowing method is designed to jump out of local optimum. In order to demonstrate the validation of DA, we compare the proposed algorithm with other existing algorithms, including bat algorithm, particle swarm optimization, and enhanced fireworks algorithm. Simulations show that the proposed algorithm seems much superior to other algorithms. At the same time, the proposed algorithm can be applied to optimize extreme learning machine (ELM) for biomedical classification problems, and the effect is considerable. At last, we use different fusion methods to form different fusion classifiers, and the fusion classifiers can achieve higher accuracy and better stability to some extent.
TL;DR: This work proposes an Advanced ELM Ensemble (AELME) for classification, which includes Regularized-ELM, L2-norm-optimized ELM (ELML2), and Kernel- ELM, and employs an objective function of increasing diversity and accuracy among the final ensemble.
Abstract: Extreme Learning Machine (ELM) is a fast-learning algorithm for a single-hidden layer feedforward neural network (SLFN). It often has good generalization performance. However, there are chances that it might overfit the training data due to having more hidden nodes than needed. To address the generalization performance, we use a heterogeneous ensemble approach. We propose an Advanced ELM Ensemble (AELME) for classification, which includes Regularized-ELM, L2-norm-optimized ELM (ELML2), and Kernel-ELM. The ensemble is constructed by training a randomly chosen ELM classifier on a subset of training data selected through random resampling. The proposed AELM-Ensemble is evolved by employing an objective function of increasing diversity and accuracy among the final ensemble. Finally, the class label of unseen data is predicted using majority vote approach. Splitting the training data into subsets and incorporation of heterogeneous ELM classifiers result in higher prediction accuracy, better generalization, and a lower number of base classifiers, as compared to other models (Adaboost, Bagging, Dynamic ELM ensemble, data splitting ELM ensemble, and ELM ensemble). The validity of AELME is confirmed through classification on several real-world benchmark datasets.
TL;DR: A new method combining Biomimetic Pattern Recognition (BPR) with CNNs is proposed for image classification that performs class recognition by a union of geometrical cover sets in a high-dimensional feature space and therefore can overcome some disadvantages of traditional pattern recognition.
Abstract: As a typical deep-learning model, Convolutional Neural Networks (CNNs) can be exploited to automatically extract features from images using the hierarchical structure inspired by mammalian visual system. For image classification tasks, traditional CNN models employ the softmax function for classification. However, owing to the limited capacity of the softmax function, there are some shortcomings of traditional CNN models in image classification. To deal with this problem, a new method combining Biomimetic Pattern Recognition (BPR) with CNNs is proposed for image classification. BPR performs class recognition by a union of geometrical cover sets in a high-dimensional feature space and therefore can overcome some disadvantages of traditional pattern recognition. The proposed method is evaluated on three famous image classification benchmarks, that is, MNIST, AR, and CIFAR-10. The classification accuracies of the proposed method for the three datasets are 99.01%, 98.40%, and 87.11%, respectively, which are much higher in comparison with the other four methods in most cases.
TL;DR: The seasonal trend autoregressive integrated moving averages with dendritic neural network model (SA-D model) is proposed to perform the tourism demand forecasting and achieved good predictive performances in terms of the normalized mean square error, absolute percentage of error, and correlation coefficient.
Abstract: With the impact of global internationalization, tourism economy has also been a rapid development. The increasing interest aroused by more advanced forecasting methods leads us to innovate forecasting methods. In this paper, the seasonal trend autoregressive integrated moving averages with dendritic neural network model (SA-D model) is proposed to perform the tourism demand forecasting. First, we use the seasonal trend autoregressive integrated moving averages model (SARIMA model) to exclude the long-term linear trend and then train the residual data by the dendritic neural network model and make a short-term prediction. As the result showed in this paper, the SA-D model can achieve considerably better predictive performances. In order to demonstrate the effectiveness of the SA-D model, we also use the data that other authors used in the other models and compare the results. It also proved that the SA-D model achieved good predictive performances in terms of the normalized mean square error, absolute percentage of error, and correlation coefficient.
TL;DR: This work proposes a metaheuristic technique called Directed Bee Colony Optimization Algorithm using the Modified Nelder-Mead Method for solving the Nurse Rostering Problem.
Abstract: The Nurse Rostering Problem is an NP-hard combinatorial optimization, scheduling problem for assigning a set of nurses to shifts per day by considering both hard and soft constraints. A novel metaheuristic technique is required for solving Nurse Rostering Problem (NRP). This work proposes a metaheuristic technique called Directed Bee Colony Optimization Algorithm using the Modified Nelder-Mead Method for solving the NRP. To solve the NRP, the authors used a multiobjective mathematical programming model and proposed a methodology for the adaptation of a Multiobjective Directed Bee Colony Optimization (MODBCO). MODBCO is used successfully for solving the multiobjective problem of optimizing the scheduling problems. This MODBCO is an integration of deterministic local search, multiagent particle system environment, and honey bee decision-making process. The performance of the algorithm is assessed using the standard dataset INRC2010, and it reflects many real-world cases which vary in size and complexity. The experimental analysis uses statistical tools to show the uniqueness of the algorithm on assessment criteria.