Search

Various machine learning algorithms are developed with an aim to create precise and trustworthy models and extract knowledge from data sources. Deep expertise in the field of machine learning is required for the challenging task of choosing the right algorithms for a specific dataset. There is no single algorithm that outperforms all others across all applications and different datasets. The difficulty of choosing an appropriate algorithm for a specific task in specific domain is related to the properties of the dataset. Properties of the dataset are measured through meta-features. Meta-features describe task and can provide explanation how one machine learning approach outperforms other algorithms on a given dataset. Learning about the effectiveness of learning algorithms, or meta-learning was developed to deal with this issue. Focus is required because previous research papers have not successfully identified meta-features in particular domains. In this research, we have evaluated various meta-feature characterization methodologies and have concentrated on basic meta-features. Business domain data is in the focus of this paper. We computed basic (general) meta-features and illustrated several use cases for their applications. [ABSTRACT FROM AUTHOR]

The sine and cosine algorithm has become a widely researched swarm optimization method in recent years due to its simplicity and effectiveness. Based on the advantages, the study in this paper delves deeper into the key parameters that influence the performance of the algorithm, and has implemented modifications such as integrating the reverse learning algorithm and adding elite opposition solution to create the modified Sine and Cosine Algorithm (the modified SCA). Furthermore, by combining the fuzzy k-nearest neighbor method with the modified SCA, the study simulates numeric datasets with two or multiple classes, and analyzes the results. The accuracy rate (ACC) achieved by the modified SCA FKNN in this paper is compared to other models, with data comparison results and tables presented for each. The modified SCA FKNN proposed in this paper has obvious advantages on accuracy rate(ACC). [ABSTRACT FROM AUTHOR]

Artificial Intelligence (AI) is an essential component in modern games. With AI, players can feel the challenges in the game and the game will feel more real. AI has several branches, one of which is path-finding. Pathfinding is a way of finding the shortest path between two points. The main problem in path-finding is how to perform a path-finding accurately and requires fewer computational resources (CPU and memory). This paper describes the results of research that tested the A* and Hierarchical Pathfinding A* algorithms using the Unity 3D platform and the C# programming language. The graph or space to be used is a graph with 8 branch nodes. While the benchmarks used are the path generated and the processing time of an algorithm. This paper results in a conclusion that the path generated by the A* algorithm is shorter than the Hierarchical Pathfinding A* algorithm. The number of paths processed for the A* algorithm is more than the Hierarchical Pathfinding A* algorithm. The total execution time of the A* Algorithm is smaller than that of the A* Hierarchical Pathfinding Algorithm. The Hierarchical Pathfinding A* algorithm experiences spikes in execution time more often than the A* algorithm. However, if the total spike time, the Hierarchical Algorithm is less than the A* Algorithm. [ABSTRACT FROM AUTHOR]

Twitter is known to be one of the familiar social networking platform these days, among many others, with a lot of user engagement. This microblogging site encourages social interactions, allowing users to stay up to date on the latest news and events and share them with others in real time. Tweets are limited to 280 characters and is allowed to include links to related websites and tools. With a platform having such wide reach, it is prone to be targeted negatively and spams are one way to do it. Spammers use this platform to display malicious content that is inappropriate and harmful to users worldwide. Machine Learning uses various approaches that can be used to detect spam and overcome it. However, with the advent of recent technologies it has been observed that the properties of tweets vary overtime making it difficult to detect spam leading to the "Twitter Spam Drift" problem. This paper reviews the papers published since 2018 that have focused on the spam drift problem and gives a comparative analysis of the different algorithms that are utilized on the various data sets to tackle such a problem. [ABSTRACT FROM AUTHOR]

Aim: Cloud computing has become one of the most popular storage and computing technologies. The aim of this paper is to provide a secure and cost effective platform for the employee management system. Materials and Methods: This study contains a total of 2 groups, the El Gamal encryption algorithm is analysed in the group 1 (434 samples), and the advanced encryption standard algorithm is analysed in the group 2 (434 samples). The accuracy and efficiency of each of the models are compared. Result: This paper is to improve the efficiency of the algorithm used in the employee management system. The proposed model appears to be efficient and fastest than the existing algorithm and the mean accuracy of detection is ±1SD with power value of 80%. Conclusion: The final outcome of the existing system, El Gamal encryption algorithm is compared with the outcome of the advanced encryption standard algorithm, and the proposed model proved to be significantly higher efficiency(1.3 times faster) than the existing model. [ABSTRACT FROM AUTHOR]

In the world of modern era we find many techniques for enhancement of reliability in distribution networks and meeting the consumer demand by provision of quality and incessant power supply. Amidst the various techniques, in this study, we opt for demand response (DR) and charging/discharging of electrical vehicles (EV's) are evaluated to ameliorating reliability for radially distributed power system by adopting teaching learning based optimization (TLBO) algorithm. This analysis is considered due to the contribution of technologies like are DR and EV's for reducing the usage of conventional sources and enhancing the air quality. Here in this paper we are adopting reliability evaluation by considering energy expected not supplied (EENS) and loss of load expected (LOLE)where the characteristics of distribution network contains load and voltage profile. This paper provides a new optimization approach for finding the optimal solution. [ABSTRACT FROM AUTHOR]

This paper focuses on finding the best classification algorithm model in the case study of student performance prediction and comparing the algorithm performance before and after using the normalization method. To achieve this goal, this study uses data mining classification techniques to analyse student performance at Vocational High School in 2020-2021. The steps of the research carried out include: data collection, data pre-processing, build algorithm models without using normalization and with using normalization, and final step are comparing algorithm performance before and after using normalization. The algorithms that will be used include: Random Forest, Decision Tree, Logistic Regression, SVM, Naive Bayes, and KNN. While the normalization methods used are Standard Scaler, Min-Max Scaler, and Robust Scaler. The result of this research is that the normalization method is able to significantly increase the accuracy of the model. Based on the tests and evaluations carried out, the normalization method using the Min-Max Scaler has the biggest impact in improving the overall model performance and the algorithm with the best performance is Random Forest. This paper reviews the effect of the normalization method to improve algorithm performance in predicting student performance, where based on previous research no one has used the normalization method to gain accuracy of the model which actually has a considerable impact on gaining accuracy of the model. [ABSTRACT FROM AUTHOR]

The paper examines the application of systematic and risk-based approaches to problem solving in the agro-industrial complex, which generally prompted research in the field of various model-based methodologies. Of particular interest to this study is the Quality by Design initiative in the agro-industrial complex. Motivated by their need, this paper reports some of the contributions to analysing the existence and availability of the experimental plan. The analysis of the feasibility and adaptation of the nested sampling algorithm for the probabilistic characteristics of the design space is considered. There is described an initial adaptation of the nested sampling algorithm, common for large Bayesian calculations, for the probabilistic characterization of spatial design, which is a key work for practitioners in the agro-industrial complex. The choice-based approach has been found to be effective with the optimization approach, allowing practitioners to take advantage of the choice-based approach with other methods. A step-by-step technology is given for complex and one-time implementations of the original nested sample for spatial design, which further reduces the computational load, thereby allowing solutions of more complex problems. Design-centring methodology is shown as an alternative coding method for methodological choice, providing the practitioner with a convenient format for communicating with process operators. A particular attention is paid to several pronounced problems that arise in the special design of experiments when there are various levels of model uncertainty, which are often encountered in the early stages of model development. The paper presents the results of the study obtained by developing optimal experimental queries in the presence of restrictions. [ABSTRACT FROM AUTHOR]

The paper describes selection of the processor architecture for the external control block implementation in assembler. This block is used to control the operation of the main control system and improve the reliability of the system as a whole. A scheme for the considered method implementation with an external control unit on CPU instead of FPGA is proposed. The development of software for it is realized in assembler. Analysis of the architectures' peculiarities that are currently available on the market is carried out. They are x86 and x86-64 for CISC. ARM v8, ARM64, RISC-V, SPARC, and LEON are for RISC. The paper performs translation of decision-making algorithms in multiversion environments from the C++ language into the assembler of the necessary architectures. The size of the resulting code is analyzed. The authors made conclusions on the basis of the obtained results. It is preferable to select a CISC architecture, since it simplifies and speeds up a software development process in the case when it is required to develop a code in assembler. As for the scheme selection for the control system implementation with an external control unit, the authors prove that it is quite acceptable to apply a CPU for the external unit instead of the initially proposed FPGA. [ABSTRACT FROM AUTHOR]

Blood sugar levels rise as a result of impaired glucose homeostasis, which is a feature of Diabetes Mellitus (DM). Coronary artery disease, nerve damage, diabetic retinal disease, kidney failure, and sex disorders are just a few of the consequences that can result from DM. DM has to be diagnosed early to minimize its effects and those of its consequences. One technique for automatically detecting DM based on the prior data is the clustering algorithm. However, the K-mean Clustering method is easy to be implemented, has fast computational time and adapted easily. However, the centroids of K-means are initialized randomly that causes to be stuck in local optima. For this reason, the robust centroids initialization on the K-means clustering method are necessary to handle this problem. This paper uses swarm intelligence algorithm for this goal since it is able to obtain solution near the global optima, is robust and easy to be implemented. As a result, this study proposes an improved K-means clustering-based technique for diagnosing DM. The K-means is improved by using robust centroid initialization-based swarm intelligence algorithm. The swarm intelligence algorithm which is used is Particle Swarm Optimization. By using experiments, the proposed method is confirmed better than the original K-means clustering in diagnosing DM. [ABSTRACT FROM AUTHOR]

This paper describes an approach to improving the process of resource redistribution in a coevolutionary algorithm using a fuzzy logic system. Two bases of fuzzy rules are proposed for setting the size of the fine and the social card for the losing coevolution algorithms. The effectiveness of the developed fuzzy system was studied in comparison with the usual co-evolutionary algorithm without integration of fuzzy control procedures. [ABSTRACT FROM AUTHOR]

The paper presents an analysis and review of sources considering temperature-dynamic characteristics as performance indicators of agricultural automotive and tractor equipment, the impact of energy efficiency of automotive equipment on the temperature and dynamic characteristics of cooling systems, an analysis of the impact of performance indicators of automotive equipment on the efficiency of a block cooling system. An improved scheme of a block-modular cooling system and a method for determining its main indicators during operation for functional units of automotive equipment using innovative materials are proposed. A complex method was used, including an analysis of the state of the issue, theoretical studies of the operating parameters of the polymer radiator of the cooling system with the inclusion of thermal, aerodynamic and hydrodynamic calculations, both in steady and unsteady modes of operation of the MTZ-82 tractor engine. At the same time, analytical and numerical methods, a set of mathematical models and graphic visualization of solutions using the Excel, KOMPAS-3D, Statistica software packages were used. An improved theory of temperature-dynamic characteristics of automotive equipment is used in the cooling system of the MTZ-82 tractor equipped with a polymer radiator, while combined options for using both liquid and oil-liquid engine cooling are possible. [ABSTRACT FROM AUTHOR]

In this paper, logical controllers, their scope of application as a type of intelligent control and classification were considered. Precise logic controllers with a step-time algorithm for fuzzification of input parameters were also considered as an attempt to eliminate the main disadvantages of logic controllers with an anytime-algorithm. A logical scheme was given that implements the step-time algorithm for fuzzification of a continuous physical quantity and an analysis of expressions was carried out to increase the speed of the fuzzifier and controller with the step-time algorithm, respectively, due to the development of a system of production rules to a rule whose antecedent at the current time is equal to a logical unit, from which it is concluded that the speed of precise logical controllers with step-time algorithm has increased due to the lack of a search procedure in the system of production rules, rules with an antecedent, equal at a given time to a logical unit. Although in case of an unsuccessful arrangement of the rules in the fuzzifier production system (the rule with the highest response frequency is located at the end of the production system), the increase in the speed of the controller may be insignificant. [ABSTRACT FROM AUTHOR]

This paper solves the optimal reactive power dispatch (ORPD) problem applying sine-cosine optimization algorithm. The main intention of the ORPD problem is to reduce the real power loss in the transmission lines and maintain voltage magnitude at an acceptable level in the system network while satisfying the set of specified operational constraints. The sine-cosine algorithm is a powerful population formed algorithm for solving the optimization problems. The sine-cosine algorithm generates a number of initial candidate solutions, and progresses towards the optimal solution. The proposed method is applied to standard IEEE-14 and 30 bus systems. The simulation results compare favorably with other reported results in the literature. [ABSTRACT FROM AUTHOR]

Given the importance of transportation in logistics operations, the vehicle routing problem (VRP) is critical to logistics practice. The Green Vehicle Routing Problem (GVRP) is a variation of the VRP that focuses on issues related to sustainable transportation. GVRP's goal is to balance the economy and environmental costs by developing effective solutions to environmental and financial concerns. The purpose of this paper is to propose a Salp Swarm Algorithm (SSA) to solve the GVRP. Using the benchmark test instance E-n23-k3, a numerical experiment is carried out to test the proposed algorithm. The results show that the SSA outperforms the Route Splitting Tabu Search (RS-TS). On average, the total cost is reduced by 1.41%. Furthermore, we show that the use of various modes of transportation affects fuel consumption, emission costs, and vehicle usage costs. [ABSTRACT FROM AUTHOR]

Activity recognition offers a wide range of applications in various industrial processes and healthcare. This work proposes an approach to collect data from a spherical coordinate system using smartphones, then extract the highly efficient features using advanced preprocessing. The paper also proposes an algorithm to recognize activity using various ensemble machine-learning approaches based on extracted features. These approaches are evaluated under various combinations of features to analyze the accuracy, sensitivity, specificity, and training time. Experimental results reveal that weighted KNN performs best among all models by achieving 96.2% accuracy with 12 features. On the other hand, Bagged tree ensemble classifiers perform better than subspace KNN ensemble classifiers with an accuracy of 95.3% using 12 features. [ABSTRACT FROM AUTHOR]

The paper analyzes the influence of the adaptive mechatronic system parameters on the process of stabilizing the step of moving materials during automatic assembly. Dependences of variances and mathematical expectations of the displacement step are obtained depending on the parameters of the system under a random external action. Numerical simulation of the operation of the algorithms used is carried out, recommendations are given for the optimal choice of system parameters. [ABSTRACT FROM AUTHOR]

This paper describes in detail the method and algorithm for calculating the numerical integration of non-stationary two-dimensional systems of Navier-Stokes equations for a compressible gas using high-order implicit (explicit) difference schemes, i.e., the Beam-Warming difference scheme. A number of mathematical transformations, such as non-dimensionalization of coordinates and physical parameters, transforming the shape of the channel into a square one, as well as thickening the integration steps with large gradients of unknowns, which make it possible to bring problems to a model one for solving similar problems. In addition to the study, as convergence in the number of calculated points, the dependence of the number of iterations on the Courant number and comparison of the results obtained by different proposed models for calculating the effective viscosity, as well as a number of numerical studies as co-occurrence, temperature inhomogeneities, non-design, the ratio of the sizes of the inlet slots of cocurrent flows and the length of the channel to the height of the main flow and the degree of expansion of the channel to the process of mixing and distribution of flows in the channel. It was revealed at what ratios of the initial parameters of the cocurrent flows and the linear dimensions of the channel a recirculation zone is formed. For a channel of constant cross section at large ratios of velocities (72, 46) and temperature (2, 33), as well as small ratios of the channel length to the height of the inlet slot of the main flow (5, 31), even at the same jet pressures in the initial section, the recirculation zone occupies more than 55% of the input section, and the length along the longitudinal coordinate reaches 20 cm. [ABSTRACT FROM AUTHOR]

In the present paper algorithms for constructing a finite element mesh of compound domains and minimizing the ribbon width of the system of resolving equations by the finite element method (FEM) are considered. A technology for constructing a finite element mesh by connecting elementary subdomains, a semi-automatic method for partitioning a two-dimensional multiply connected domain with a curvilinear boundary, and ordering the node numbers of a finite element mesh is described. In this case, the algorithms allow, by ordering the numbering of nodes, to determine such a sequence in which the ribbon width of non-zero coefficients of the system of resolving equations by the FEM will be locally minimal. [ABSTRACT FROM AUTHOR]

Feature selection is an important technique in building prediction systems. In various searches, the judgment of the relevancy of a feature is often calculated using all the instances of the considered sample. However, when the dataset size grows, some of the instances are not useful for weighting the features. In this paper, we rank features according to a fitness function that relies on the relevancy using all instances and on the relevancy using a maximum number of significant instances (which really contribute to the feature positive relevancy with the target variable). The relevancy is based on preordonnances theory where the instances are expressed in pairs. The proposed algorithm can be mainly divided into three steps, namely, (a) eliminating all features that are in disagree with the target feature, (b) Finding the best subset of instances, to each feature, that maximize the relevancy and which the cardinal tends to the cardinal of instances in the original dataset, and (c) Ranking features. The second step is defined by dividing the dataset (instances of each feature) into several consistent regions by fuzzy clustering. Then, performing GA-based instances selection independently within each cluster. Finally, aggregating of the partial results by the ensemble voting. Experimental results verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Uncertainties, parameters variations and external disturbances may affect highly the task of lane-keeping, lane-changing and path-tacking autonomous vehicles. This paper presents a novel variable gains second-order sliding mode algorithm for en-hanced path tracking performance of autonomous vehicles subject to uncertainties and disturbances. The used algorithm is a modified super-twisting with both nonlinear and linear correction terms. Hence, the proposed method doesn't suffer from the chattering phenomenon, ensures high order precision and guarantees both the robustness against perturbations and the convergence in finite-time during the reaching phase. To evaluate the efficiency of the developed approach, a double lane-changing scenario is considered in simulation for an uncertain autonomous vehicle and the results are compared with the first-order sliding mode. [ABSTRACT FROM AUTHOR]

The concept of web of things (WoT) is to connect anything in the physical world and then display them in the World Wide Web, that is the major objective, but that mean information storage or transfer that led to thinking how can protected this information from attacks, this can happen by protected information by encryption it. Cryptography and chaos inextricably linked because of many features. In this paper uses a novel model for the encryption of information via WoT, we suggest HIGHT and Simon as a block cipher algorithm using for lightweight encryption by make a new hybrid between them. Also use chaotic system to take key from it; moreover, we use other operations to arrive the best result. [ABSTRACT FROM AUTHOR]

Path planning algorithms is the most significant area in the robotics field. Path Planning (PP) can be defined as the process of determining the most appropriate navigation path before a mobile robot moves. The goal of optimization of PP is to find a path that is close to or optimal path. Multi-objective optimization (MOO) is concerned with finding the best solution values that satisfy multiple objectives, such as shortness, smoothness, and safety. MOO present the challenge of making decisions while balancing these contradictory issues through compromise (tradeoff). As a result, there is no single solution appropriate for all purposes in MOO, but rather a range of solutions. This paper the problem of PP is investigated with path length, security and smoothness objectives for autonomous mobile robots in dynamic environment with moving target. The proposed approach based on Bat algorithm (BA). The proposed solution is tested in a number of simulations with varying scenarios. The simulation results show that multi objective can be achieved efficiently in a dynamic environment which proves its effectiveness. [ABSTRACT FROM AUTHOR]

During pandemic circumstance world, the size of data is exploded like rocket speed. At the same time, there is a need to process it and investigate the enormous volumes of information. The company and research tool is Hadoop and is most broadly utilized in information serious applications. The Hadoop default processing nature is homogeneous. The limit of enormous measure of data in Hadoop is finished utilizing Hadoop Distributed File System. The performance is degraded when it applied to heterogeneous environment. Hadoop mainly uses 2 components viz HDFS used for storage purpose and MapReduce used for programming. In this paper we introduce Optimal Data Placement Algorithm (ODPA). [ABSTRACT FROM AUTHOR]

The real world has been changed into a virtual one as a result of the pandemic scenario. Cloud computing plays a major role for storing and computing the data using remote computing infrastructure for solving the day to day activities. The LB Techniques are used in the cloud computing process. LB techniques distributing the system's load among its various nodes in order to maximise resource efficiency and job response time. The major issues in load balancing are Geographical Distributed system, Single point of failure (SPoF), live Migration of instance, Heterogeneous device, Storage device Management, Scale-out/scale-in and Complexity of Algorithm. Load balancing ensures that each node in the network shortens reaction times, reduces expenses, and boosts performance. This paper address the various cloud computing balancing algorithm survey. [ABSTRACT FROM AUTHOR]

Today's modern world is being driven towards automation and as well gathering any type of information from every nook and corner of the world which is possible through one of the technology named Internet of Things(IoT). IoT is not just a single technology but it is a combination of various technologies like embedded systems, wireless sensor networks, cloud computing etc. Basic IoT network consists of various layers; one among them is the device layer which is the base layer responsible for gathering the data from physical world through the sensors. Various sensors can be deployed into an area depending upon the type of application it is used for. Various homogenous sensors can be made into clusters due to which it would be beneficiary in terms of bandwidth and power consumption, number of internet connections required, fault tolerance, longevity of devices etc. This paper presents a comparison between various algorithms used to cluster the sensors at device level. [ABSTRACT FROM AUTHOR]

By using thoracic electrical bioimpedance (TEB) analysis stroke volume calculated for sudden cardiac arrests. In this paper proposed robust normalized diffusion least power algorithm (RNDLPA) is proposed to get high resolution components. In clinical scenarios, TEB components are contaminated with various artifact components, which interrupts the identification of several features in stroke intensity. The number of multiplications is also one of the key elements for monitoring in health care applications. Hence by proposed RNDLPA algorithm it reduces computational complexity, stability is improved and also steady state error rate is reduced. To eliminate noise components various adaptive cancelers are presented and they are compared with proposed algorithm. Further to show performance improvement signal to noise ratio (SNR) improvement is taken into consideration it gets better values 14.3594 dB and 12.3551 dB for respiratory noise and muscle noise respectively when compared to existed techniques. [ABSTRACT FROM AUTHOR]

Xilinx System Generator (XSG) integrates with MATLAB Simulink to build a complex system through Xilinx and Simulink blocks. This paper presents a hardware implementation of various image processing techniques to modify and extract the pixel information from various input images in run time. Complex coding strategies are eradicated with integrated system. The intricacy in structural design is reduced to a huge extent using Xilinx System generator. The reconfigurable attribute of Field Programmable Gate Array (FPGA) bags the massive advantages in image processing like co simulating with selective pixel and verified on FPGA. Image processing algorithms explored in methodical manner by performing diverse operations like edge detection, and color enhancement, complement, contrast stretching etc. The proposed image processing techniques (IPT) intended for various applications. Dissimilar image processing techniques are processed by FPGA and displayed on Video Graphics Array (VGA). Image processing system is designed in XSG, hardware description language (HDL) like VHDL & Verilog code is generated by processing system by system generator token. Design is implemented on Artix-7 and targeted on Basys3 reconfigurable hardware (XC7A35T-1-CPG236C). [ABSTRACT FROM AUTHOR]

We live in times where technology is advancing at an unprecedent rate. From online communication, shopping, data sharing, cashless payments to transmitting sensitive information through internet, everything is being done online. These online methods of data transfer are always vulnerable to online phishing and frauds. Therefore, it becomes very imperative to devise ways for safe and secure transmission of data. To keep up this, new and secure cryptographic schemes are the need of the modern age. In this paper, the author presents a new improvised encryption technique which makes modifications in the traditional approach of the Caesar cipher by combining it with extended Hill cipher. Additionally, it makes use of the Linear congruential method to generate the randomized secret key. [ABSTRACT FROM AUTHOR]

The paper considers the main methods of testing a software product at the development stage. The methods are compared with typical architectures of software products. The optimization problem of the testing process is solved on the basis of the algorithm development that presents the process under consideration at a more detailed level. The optimization of the algorithm is presented on business processes by BPMN 2.0 notation. [ABSTRACT FROM AUTHOR]

The article presents the software implementation of the bioinspired ant colony algorithm, which is used to form the composition of the multiversion software modules. It is shown that the reliability of programs and performance indicators of multiversion software significantly depend on the results of the design stage, at which a decision is made on the formation of the structure of the program complex and the choice of the composition of multiversion software modules. The solution of these problems is supported by the developed software application. The application is focused on the user, who sets all the parameters of the multiversions. Within the framework of the bioinspired approach, the paper presents the main functions implemented in a software package using ant colony algorithms. [ABSTRACT FROM AUTHOR]

Active Contour Model is the most original algorithm among the class of Contour Finding methods. Its capability is tested against wound parameter detection given image data. The performance is less satisfying, in general, few perimeters are found correctly, but the rest are far from ideal. However, through some tuning by adding an extra pre-processing step, the result can be improved. Image Interpolation is used to achieve this improvement. This paper reports how significant the improvement is. [ABSTRACT FROM AUTHOR]

One of the measurements for evaluating time series forecasting performances is the mean square error (MSE). This paper proposes an algorithm to find the smallest MSE. To capture the four components of the time series data (namely, seasonal variation, trend variation, cyclical variation, and random variation), the exponential smoothing method was used. This method uses three factors for smoothing where the data are the smoothing factor, 0 < α < 1, the trend smoothing factor, 0 < γ < 1, and the seasonal change smoothing factor, 0 < β < 1. First, all possible combination values of smoothing factors will be generated to one decimal digit. After that, the smallest MSE of those combinations will be determined by using an algorithm and observing their convergence pattern. [ABSTRACT FROM AUTHOR]

In this paper, we propose a modified mean-variance portfolio selection procedure using the proximal gradient method. This novel selection procedure has the advantage of encouraging sparsity in the entire portfolio usingl0−norm while requiring only O(n) storage. The proximal gradient method (PSG) introduces a multiple damping gradient (MDG) method to reduce the computation time, with Lipschitz constant serving as the step size, and iterative thresholding (IHT) as the proximal method to induce the sparsity of the portfolio. To find the optimal portfolio, an efficient algorithm is developed. The performance of the proposed selection procedure is illustrated by its application to real-life data set. It is shown that the proposed portfolio performs consistently better in comparison with other methods. [ABSTRACT FROM AUTHOR]

In this paper we build a constructive algorithm that returns a rectifiable curve that connects two points in a weakly convex set in a Hilbert space. We have proven that this algorithm converges and obtained an estimate on the curve's length and compare the length of the curve obtained to known results. [ABSTRACT FROM AUTHOR]