Virtual machine placement (VMP) and energy efficiency are significant topics in cloud computing research. In this paper, evolutionary is applied to VMP minimize the number of active physical servers, so as schedule underutilized servers save energy. Inspired by promising performance ant colony system (ACS) algorithm for combinatorial problems, an ACS-based approach developed achieve goal. Coupled with order exchange migration (OEM) local search techniques, resultant termed OEMACS. It...

In pedagogy, teachers usually separate mixed-level students into different levels, treat them differently and teach in accordance with their cognitive learning abilities. Inspired from this idea, we consider particles the swarm as propose a level-based optimizer (LLSO) to settle large-scale optimization, which is still considerably challenging evolutionary computation. At first, strategy introduced, separates number of levels according fitness values treats differently. Then, new exemplar...

Large-scale optimization has become a significant yet challenging area in evolutionary computation. To solve this problem, paper proposes novel segment-based predominant learning swarm optimizer (SPLSO) through letting several particles guide the of particle. First, strategy is proposed to randomly divide whole dimensions into segments. During update, variables different segments are evolved by from exemplars while ones same segment exemplar. Second, accelerate search speed and enhance...

Keyframe selection has been crucial for effective and efficient video content analysis. While most of the existing approaches represent individual frames with global features, we, first time, propose a keypoint-based framework to address keyframe problem so that local features can be employed in selecting keyframes. In general, selected keyframes should both representative containing minimum redundancy. Therefore, we introduce two criteria, coverage redundancy, based on keypoint matching...

Surrogate-assisted evolutionary algorithms (SAEAs) have become one popular method to solve complex and computationally expensive optimization problems. However, most existing SAEAs suffer from performance degradation with the dimensionality increasing. To this issue, article proposes a classifier-assisted level-based learning swarm optimizer on basis of (LLSO) gradient boosting classifier (GBC) improve robustness scalability SAEAs. Particularly, strategy in LLSO has tight correspondence...

In tackling data mining and pattern recognition tasks, finding a compact but effective set of features has often been found to be crucial step in the overall problem-solving process. this paper, we present an empirical study on feature analysis for classical instrument, using machine learning techniques select evaluate extracted from number different schemes. It is revealed that there significant redundancy between within schemes commonly used practice. Our results suggest further research...

Automated detection of motorcycle helmet use through video surveillance can facilitate efficient education and enforcement campaigns that increase road safety. However, existing approaches have a number shortcomings, such as the inabilities to track individual motorcycles multiple frames, or distinguish drivers from passengers in use. Furthermore, datasets used develop are limited terms traffic environments density variations. In this paper, we propose CNN-based multi-task learning (MTL)...

Wireless sensor network (WSN) technologies have been employed in recent years for monitoring purposes various domains from engineering industry to our home environment due their ability intelligently monitor remote locations. In this paper, we developed a purely deterministic model that utilizes clustering organize the WSN. We propose energy-efficient protocol is dynamic, distributive, self-organizing and more energy efficient than existing protocols. It simplified approach which minimizes...

In order to support smart construction, digital twin has been a well‐recognized concept for virtually representing the physical facility. It is equally important recognize human actions and movement of construction equipment in virtual scenes. Compared extensive research on action recognition (HAR) that can be applied identify workers, field (CEAR) very limited, mainly due lack available datasets with videos showing equipment. The contributions this are as follows: (1) development...

Capturing individual differences in cognition is central to human neuroscience. Yet our ability estimate cognitive abilities via brain MRI still poor both prediction and reliability. Our study tested if this inability can be improved by integrating signals across the whole modalities, including task-based functional (tfMRI) of different tasks along with other non-task such as structural MRI, resting-state connectivity. Using Human Connectome Project (n = 873, 473 females, after quality...

The construction sector is widely recognized as having the most hazardous working environment among various business sectors, and many research studies have focused on injury prevention strategies for use sites. risk-based theory emphasizes analysis of accident causes extracted from reports to understand, predict, prevent occurrence accidents. first step in classify incidents a massive number into different cause categories, task which usually performed manual basis by domain experts....

Diagnosing pain in research and clinical practices still relies on self-report. This study aims to develop an automatic approach that works resting-state raw EEG data for chronic knee prediction. A new feature selection algorithm called "modified Sequential Floating Forward Selection" (mSFFS) is proposed. The improved scheme can better avoid local minima andexplore alternative search routes. obtained by mSFFS displays class separability as indicated the Bhattacharyya distance measures...

Cognitive dysfunction often co-occurs with psychopathology. Advances in neuroimaging and machine learning have led to neural indicators that predict individual differences cognition reasonable performance. We examined whether these explain the relationship between mental health UK Biobank cohort (n > 14000). Using learning, we quantified covariation general 133 indices derived of from 72 phenotypes across diffusion-weighted MRI (dwMRI), resting-state functional (rsMRI), structural (sMRI)....

An algorithm of evolving self-organizing map (ESOM) is proposed as a dynamic version the Kohonen map, where network structure evolved in an online adaptive mode. Experiments have been carried out on some benchmark data sets well macroeconomic data. Results show that ESOM good tool for clustering, analysis, and visualisation.

ABSTRACT A recent research work pointed out that the reversible data hiding algorithms proposed for gray‐scale images can be implemented on reconstructed palette to improve embedding capacity and visual quality by reordering color table. However, effect has a significant impact performance improvement. Therefore, we propose clustering‐based method in further enhance performance. In this method, first design centroid initialization select initial centroids then exploit K‐means algorithm...

While wireless sensor networks (WSN) are increasingly equipped to handle more complex functions, in-network processing still requires the battery-powered sensors judiciously use their constrained energy so as prolong elective network life time. There a few protocols using clusters coordinate consumption in WSN, but how deal with heterogeneity remains research question. The authors propose modified clustering algorithm three-tier setting, where among nodes is adaptive levels. A theoretical...

Data reduction strategy is one of the schemes employed to extend network lifetime. In this paper we present an implementation a light-weight forecasting algorithm for sensed data which saves packet transmission in network. The proposed Naive achieves high energy savings with limited computational overhead on node. Simulation results from realistic Building monitoring application WSN are compared well-known prediction algorithms such as ARIMA, LMS and WMA models. We implemented real-world...

The popular performance profiles and data for benchmarking deterministic optimization algorithms are extended to benchmark stochastic global problems. A general confidence interval is employed replace the significance test, which in traditional methods but suffering more criticisms. Through computing bounds of visualizing them with (or) profiles, our method can be used compare by graphs. Compared methods, synthetic statistically therefore suitable large sets some sample-mean-based e.g.,...

Semantic segmentation based on deep learning methods can attain appealing accuracy provided large amounts of annotated samples. However, it remains a challenging task when only limited labelled data are available, which is especially common in medical imaging. In this paper, we propose to use Leaking GAN, GAN-based semi-supervised architecture for retina vessel semantic segmentation. Our key idea pollute the discriminator by leaking information from generator. This leads more moderate...