International Journal of Computing Science and Mathematics (76 papers in press)

Regular Issues

• Reformulation of Bilevel Linear Fractional/Linear Programming Problem into a Mixed Integer Programming Problem Via Complementarity Problem  
  by Anuradha Sharma 
  Abstract: The bilevel programming problem is a static version of the Stackelberg's leader follower game in which in which Stackelberg strategy is used by the higher level decision maker called the leader given the rational reaction of the lower level decision maker called the follower.The bilevel programming problem is a two level hierarachical optimization problem and is non-convex.This paper deals with finding links between the bilevel linear fractional/linear programming problem(BF/LP), the generalized linear fractinal complememtarity problem(GFCP) and mixed integer linear fractional programming problem (MIFP),The (BFLP) is reformulated as a(GFCP) which in turn is reformulated as an (MIFP),The method is supported with the help of a numerical example.
  Keywords: Bilevel Programming; Generalized complementarity problem; Mixed integer programming; Fractional programming.
  
  
• Succulent Link Selection Strategy for underwater sensor network  
  by Shahzad Ashraf 
  Abstract: In underwater environment, the sensor nodes are deployed for collecting information and sending back to the base station. Establishing astute communication link among these sensor nodes in a multi-link routing environment is a key challenge for all underwater routing protocols. A sagacious communication link can only guarantee the maximum data transfer rate. The link selection mechanism of three underwater routing protocol i.e, Energy-aware Opportunistic Routing (EnOR) protocol, Shrewd Underwater Routing Synergy using Porous Energy Shell (SURS-PES) and Underwater Shrewd Packet Flooding Mechanism (USPF) have been investigated. After analyzing performance results of these protocols interms of packet delivery ratio, end-to-end ?delay, network lifespan and energy consumption using NS2 with AquaSim 2.0 simulator. The protocol existing, with sagacious link selection mechanism in multi-link routing environment has been identified. The identification of this sagacious link selection mechanism is a novel approach which can give specific knowledge for targeted output without wasting resources for irrelevant objectives.
  Keywords: Flooding mechanism; underwater routing; link selection; sagacious link; sink node.
  
  
• Combination of a 2D-RCA model and ANNs for texture image segmentation  
  by Assia Ayache, Soumia Kharfouchi, Fouad Rahmani 
  Abstract: Image segmentation is defined as the partition of an array of measurements taken on an image on the basis of homogeneity. In this paper, a region growing technique is used to achieve image segmentation by merging some starting points or internal small areas if they are homogeneous according to a measurement of a local region property. A 2D random coefficients autoregressive model (2D RCA) is fitted in order to identify the different textures present in the image. First, an estimation procedure using a generalized method of moments (GMM) technique is proposed to extract some local region properties. For this, a gradient-based neural network (GNN) is used to estimate the 2D RCA model parameters from a given texture. The cost function of the proposed (GNN) is based on a strong matching of the statistical moments of the corresponding 2D-RCA model and the sample moments of population image data. Experimental results demonstrate the effectiveness and the relevance of the proposed method.
  Keywords: Image segmentation; 2D RCA models; ANNs; GMM.
  
  
• A Decision Model to Improve the Performance of Inventory Management for Deteriorating Items Considering Temperature  
  by Shilpy Tayal, Neeraj Dhiman, S.R. Singh 
  Abstract: Here an inventory system for time and temperature dependent deterioration with constant demand rate has been developed. If we discuss about a cold storage warehouse then the items stored maintain its quality in a particular range of temperature. A temperature value higher or lower than this range results in an increased rate of deterioration. Considering all the possible cases of temperature the optimal value of unit time profit have been calculated at different points in the assumed range. With the help of numerical example it is concluded that unit time profit is maximum at that temperature value which is feasible for that particular item. With numerical analysis the optimality of the system in all the three cases has also shown graphically. Further to verify the system stability sensitivity analysis has been performed and the system is found to be quite stable.
  Keywords: Inventory; Temperature and Time Dependent Deterioration; Demand; Shortages; Partial Backlogging.
  
  
• Application of New Entropy Measure for Pythagorean Intuitionistic Fuzzy Sets  
  by Taruna , H.D. Arora, Pratiksha Tiwari 
  Abstract: Pythagorean fuzzy set (PFS) theory has been commonly used to tackle vague information and imprecise expressions in real-world decision-making contexts. In this paper, we have explained the concept of the probability of the characteristics of a fuzzy set in the Pythagorean as an extension of the fuzzy set of intuitions. In contrast to past measures, the new proposed measures are simpler, closer to the real meaning, and better reflective properties. New generalized parametric measures of exponential entropy on intuitionistic fuzzy sets are characterized. Further, fundamental properties are demonstrated. Through dissecting the consequences of the model, it has been demonstrated that this technique is progressively dependable and increasingly e?ective in handy applications for introducing the level of fuzziness of IFS. Numerical illustration is revealed to validate the methods and compare their effectiveness with existing IFS measures.
  Keywords: fuzzy entropy; intuitionistic fuzzy sets; Pythagorean fuzzy sets; linguistic variables; multi-criteria decision making.
  
  
• Modelling the impact of lock-downs in top-4 COVID-19 spreading states of India  
  by Adarsh Anand, Mohini Agarwal, Niyati Aggrawal, Navneet Bhatt 
  Abstract: Pandemics like COVID-19 being a highly infectious disease has severely affected the mankind and business activities. Seeing the critical situation, the honourable Prime Minister of India has called the lockdown in the entire country in order to supress the spread of this pandemic. While there are many debates about the spread of disease and lockdown in the entire country. We wish to mathematically understand the diffusion of this pandemic in context of four highly infected states of India. Moreover, through this article we wish to examine the impact of these lockdown period for understanding the spread-ness of COVID-19.
  Keywords: Bass Model; Coronavirus; Diffusion; Lockdown; Pandemic.
  
  
• Modified Exponential Family for Improved Searls Estimation of Finite Population Mean  
  by S.K. Yadav, Dinesh Sharma, Madhulika Dube 
  Abstract: This paper proposes a modified exponential ratio type Searls (1964) class of estimators for the estimation of population mean under simple random sampling scheme. The suggested estimator utilizes the known information on highly correlated auxiliary attribute. The theoretical derivations for the bias and mean squared error for the proposed are retained up to the approximation of order one and the performance properties of the estimator are compared with the well-established ratio, product, modified ratio and modified product estimators of the mean of the population for the characteristic under study. The efficiency conditions of the suggested estimator over these estimators are also obtained and the theoretical findings are verified using the empirical data sets. The efficiencies of the estimators are judged on the basis of mean squared errors of the sampling distributions around the true population mean of the study variable.
  Keywords: Study variable; Auxiliary Attribute; Exponential ratio estimator; Bias; MSE.
  
  
• Impacts of wind and anti-predator behaviour on predator-prey dynamics: A modelling study  
  by Prabir Panja 
  Abstract: In this paper, a predator-prey interaction model of anti-predator behaviour and wind effects has been developed. It is assumed that in the absence of predator prey growlogistically. It is also assumed that the prey shows anti-predator behaviour (group defense) against their predator. To analyze our proposed model, the impacts of wind direction have been incorporated. It is considered that due to wind effect the density of predator decrease.Next,wehave studied boundedness of all solution of the system and then all possible equilibrium points are determined. The local stability of our proposed system has been analyzed around these equilibrium points. Hopf bifurcation of our proposed system has been investigated with respect to the anti-predator behaviour (eta), wind effects (alpha) and inverse inhibitory effect of predator (b1). It is found that the effects of wind direction and anti-predator behaviour of prey can be stabilized our proposed system. Finally, some numerical simulation results have been presented to verify our theoretical findings.
  Keywords: Prey; Predator; Wind; Anti-predator behaviour; Hopf bifurcation.
  
  
• Two-Stages Explicit Schemes Based Numerical Approximations of Convection-Diffusion Equations  
  by Mohammad Izadi 
  Abstract: A wide family of finite difference methods for the convection-diffusion problems based on an explicit two-stages scheme and a seemingly implicit method are presented. In this paper, to have a greater stability region while keeping the second-order accuracy, a family of methods which combines the MacCormack and Saul'vey schemes is proposed. The stability analysis of the combined methods is investigated using the von Neumann approach. In each case, it is found that it is the convection term that limits the stability of the scheme. Based on the von Neumann analysis, valuable stability limits in terms of mesh parameters for maintaining accurate results are determined in the analytic manner and demonstrated through computer simulations. Two model problems consist of linear advection-diffusion and nonlinear viscous Burgers equation are given to illustrate some properties of the present technique such as stability and ability to propagate discontinuities.
  Keywords: Convection-diffusion equations; Finite difference approximation; MacCormack scheme; Saul'yev scheme; Stability analysis.
  
  
• Spherical based routing protocols for three-dimensional MANET  
  by Alaa Eddien Abdallah, Ebaa Fayyoumi, Emad Eddien Abdallah, Ahmad Qawasmeh, Islam Almalkawi 
  Abstract: Position-based routing protocols usually assume that mobile nodes are distributed in $2D$ space. Thus many of the previously proposed routing algorithms do not support many practical scenarios if MANET nodes are dispersed in three-dimensional environments. In this article, a couple of geographical-based routing protocols were proposed for 3D MANET, so-called, Spherical routing and Greedy-Spherical routing. In the Spherical algorithm, two dimensional Face routing is used on the internal surface of a predefined sphere after projecting each mobile node on that surface. Greedy-Spherical starts with the Greedy routing algorithm as long as there is progress toward the location of the target node. If the next node is not closer to the destination than the current node, Greedy-Spherical shifts to Spherical routing. We evaluate the newly proposed algorithms by simulation, which shows a considerable improvement in packet delivery compared with traditional known algorithms.
  Keywords: Ad hoc network; Localized routing; Delivery rate; Position based routing.
  
  
• Optimal Searls Estimation of Population Variance under Systematic Sampling Scheme: A Simulation Study  
  by S.K. Yadav, Dinesh Sharma, Abhishek Yadav, Surendra Kumar 
  Abstract: This paper proposes an improved estimation of population variance, utilizing known auxiliary information in a systematic sampling scheme. To enhance population variance estimation, we suggest a Searls (1964) type estimator utilizing known auxiliary parameters. The bias and Mean Square Error (MSE) are derived up to an approximation of first degree. The optimal values of the Searls characterizing constants are obtained, and the corresponding least mean squared errors are also obtained. The suggested estimators are theoretically compared with the competing estimators. The efficiency conditions of the suggested estimators over competing estimators are obtained. The theoretical efficiencies are verified using a real primary data set collected from a block of Barabanki District of Uttar Pradesh State in India. The estimator with lesser MSE or higher Percentage Relative Efficiency (PRE) is preferred for elevated population variance estimation in a systematic random sampling scheme.
  Keywords: Study Variable; Auxiliary Variable; Systematic Sampling; Bias; MSE; PRE.
  
  
• Solving multi-objective bi-matrix games with intuitionistic fuzzy goals through aspiration level approach  
  by Zhoushun Zheng, Mohamed Brikaa 
  Abstract: The main aim of this paper is to develop an approach to solve multi-objective bi-matrix game with intuitionistic fuzzy (IF) goals, which are called IF multi-objective bi-matrix games for short. In this article, the solution approach for such game is presented by introducing aspiration level approach, and IF non-linear programming problem is constructed to find the optimal solution for such type of multi-objective bi-matrix games. Furthermore, it is shown that this multi-objective bi-matrix game with IF goals is an extension of the multi-objective bi-matrix game with fuzzy goals. Finally, a numerical example is incorporated to demonstrate the implementation and applicability process of the proposed approach.
  Keywords: non-linear programming; intuitionistic fuzzy set; multi-objective bi-matrix games; game theory; intuitionistic fuzzy goals; aspiration level approach.
  
  
• A Support System for Coronary Artery Disease Detection using Deep Dense Neural Network  
  by Debabrata Swain 
  Abstract: Due to the development of modern gadgets and equipment, human life became very much luxurious. Hence for performing any work physical efforts are reducing day by day. This leads a person more prone to coronary artery disease which is a variety of cardiac syndrome. It has become one of the principal causes of mortality in the whole world. For the better and accurate identification of the disease, different researchers have explored many intelligent prediction systems. In this paper, an effective coronary artery disease prediction system is proposed using a deep dense neural network. The proposed model is an adaptive version of dense neural network with the addition of deep hidden layers structure and dropout. Here the data is collected from heart disease data sets present in the UCI repository. The repository consists of data taken from various geographical locations like Longbeach, Cleveland, Hungary, and Switzerland. The classifier has shown a classification accuracy of 95.32%.
  Keywords: Support Vector Machine; Random Forest; Decision Tree; Coronary Artery Disease(CAD); Deep Dense Neural Network.
  
  
• Establishment of traders optimal pricing Strategy for fresh product and used product with stock and trade cost associated demand  
  by R.P. Tripathi 
  Abstract: The recovery of used commodities is a major problem for inventory managers due to limited resources available in the universe. In this paper it is considered that a retailer sells the new commodity to purchaser and receives money, Buyer uses the product and again sells the used items. As like mango is used by living things such as men, birds, animals etc. After using it the rest part will again used for oil, medicine, fertilizers, seeds and others. Demand is assumed to be stock- sensitive for new products and price linked for used items. We formulate mathematical method for finding total yearly profit. Optimal solution is obtained by differential calculus. Numerical example and sensitivity study is providing to authenticate the authenticity of planned model. Executive phenomenon is also discussed.
  Keywords: Inventory; variable demand; recent and worn item; revenue.
  
  
• Nonparametric approximation of the characteristics of the D/G/1 queue with finite capacity.  
  by Faïrouz Afroun, Djamil Aïssani, Djamel Hamadouche 
  Abstract: In this work, we consider the finite capacity D / G /1 queue. First, the modeling of the system in question by an embedded Discrete-Time Markov chain is considered. Secondly, the aim is to illustrate the use of the discrete kernel method for the estimation of the stationary characteristics of this chain, when the general distribution that governs it is an unknown function. To support and illustrate our proposals, two extensive simulations studies are carried out.
  Keywords: Deterministic queues; Markov chains; smoothing parameter; discrete kernels; errors; simulation.
  
  
• Applications of Neutrosophic Set Theory in an Industry for Distribution of Projects and Its Maple Implementation  
  by Bizuwork Derebew, Shanmugasundaram P, Srinivasarao Thota 
  Abstract: In this paper, we proposed Average Composition Relation Method (ACRM) using the notion of neutrosophic set (NFS) operations and composition relations to identify the suitable contractors for allotment of projects. Identifying suitable contractors in any industry is not only based on bid price. Before allowing bid price, we want to ensure the prequalification criteria which affect non-price related factors like design, quality, time management, experience, financial problems. We cannot measure these factors in classical mathematics qualitatively because it is imprecise, vague and uncertain in nature. The expectations of the selection of contractors and the project criteria are recorded via Neutrosophic set and the projects allotment is effectively done in two stages. The proposed algorithm is validated through a case study. The implementation of the proposed algorithm in Maple is also discussed and sample computations are presented.
  Keywords: Neutrosophic Sets; Neutrosophic relations; Composition relations; Maple Programming.
  
  
• Forecasting foreign tourist arrivals in India using a single time series approach based on rough set theory  
  by Kriti Kumari, Haresh Sharma, Shalini Chandra, Samarjit Kar 
  Abstract: In this study, a hybrid approach based on single forecasts and rough set theory (RST) is proposed for forecasting of foreign tourist arrivals (FTAs) to India. In the formulation of the proposed hybrid method, the FTAs time series data is first forecasted using four time series models: Naive I, Naive II, Grey, and vector error correction (VEC) models. Then the RST is applied to generate an appropriate weight coefficient and the single forecasting results are combined via weight coefficient. The study also compares the forecasting results of the hybrid method with single forecasts and other combination methods such as simple average (SA) and the inverse of the mean absolute percentage error (IMAPE). Empirical results show that the proposed hybrid approach performs better than the other single forecasting modelsrn
  Keywords: Hybrid approach; Time series forecasting; tourist arrivals; single forecasts; rough Set.rn.
  
  
• A Fusion Multi-Criteria Collaborative Filtering Algorithm for Hotel Recommendations  
  by Qusai Shambour, Mosleh Abualhaj, Qasem Kharma, Faris Taweel 
  Abstract: Recommender systems employ information filtering techniques to mitigate the problem of generating personalized recommendations in the digital world that is heavily overloaded with information. Recently, tourism industry becomes more and more popular and the number of online hotel booking sites with search engines has been increasingly growing. However, using such sites can be time consuming and burdensome for potential travellers. Accordingly, this paper proposes a Fusion Multi-Criteria User-Item Collaborative Filtering Recommendation algorithm that exploits the multi-criteria ratings of users and integrates MC user-based CF and MC item-based CF techniques to produce personalized hotel recommendations. Experimental results on two real-world multi-criteria datasets show the effectiveness of the proposed algorithm by outperforming other baseline single-criteria and multi-criteria CF recommendation approaches in terms of recommendation accuracy and coverage, in particular, when dealing with sparse datasets.
  Keywords: Multi-Criteria ratings; Collaborative filtering; Recommender systems; Hotel recommendations; Sparsity.
  
  
• Solving Capacitated Vehicle Routing Problem with Route Optimization based on Equilibrium optimizer Algorithm  
  by Ibrahim Fares, Aboul Ella Hassanien, Rizk M. Rizk-Allah, R. M. Farouk, Hassan M. Abo-donia 
  Abstract: This paper presents a new solving method for the Capacitated Vehicle Routing Problem (CVRP) based on new bio-inspired Equilibrium Optimizer (EO) algorithm. The CVRP considered as one of the NP-hard combinatorial optimization problems and most of the algorithms failed to reach optimality in these problems. The EO algorithm is a powerful technique in solving several combinatorial optimization problems. The performance of the EO algorithm in solving the CVRP compared with the artificial bee colony algorithm, the particle swarm optimization algorithm, and the whale optimization algorithm. The computational results obtained for the CVRP model illustrate the power of the EO algorithm over the competitor algorithms.
  Keywords: Metaheuristic; Combinatorial optimization; Computational complexity; Natured inspired algorithms; particle swarm optimization; Artificial bee colony.
  
  
• A Mathematical Model and Optimal Control for Listeriosis Disease from Ready-to-Eat Food Products  
  by Williams Chukwu, Farai Nyabadza, Joshua Asamoah 
  Abstract: Ready-to-eat food (RTE) are foods that are intended by the producers for direct human consumption without the need for further preparation. In the present study, a deterministic model of Listeriosis disease transmission dynamics with control measures is analyzed. Equilibrium points of the model in the absence of control measures were determined, and their local stabilities established. We formulate an optimal control problem and analytically give sufficient conditions for the optimality. The transversality conditions for the model with controls are also given. Numerical simulations of the optimal control strategies were performed to illustrate the results. The numerical findings suggest that the constant implementation of joint optimal control measures throughout the modelling time will be more efficacious in controlling or reducing the Listeriosis disease. The results of this study can be used as baseline measures in controlling Listeriosis from RTE food products.
  Keywords: Listeria; Contaminated Food Products; Food Contamination Threshold; Optimal Control Interventions; Numerical Simulations.
  
  
• Pricing American put options model with application to oil options  
  by Hajar NAFIA, Yamna ACHIK, Imane AGMOUR, Asmaa IDMBAREK, Naceur ACHTAICH, Youssef EL FOUTAYENI 
  Abstract: In this paper, we reformulate a problem of pricing American put options to linear complementarity problem. The space and the time are discretized with the finite difference method in the Crank-Nickolson approach, which leads to present the put option price as a solution of the linear complementarity problem. For solving this problem and evaluating the put options we use a fast algorithm. We apply our study for an example on oil options.
  Keywords: American option; European option; Linear complementarity problem; Black and Scholes model; Crank-Nickolson approach.
  
  
• Analysis of Signed Petri Net  
  by Payal , Sangita Kansal 
  Abstract: In this paper, the behavioral properties of Signed Petri net (SPN) are given along with the two techniques: Reachability Tree and Matrix equations to analyse the SPN. An actual case scenario of a restaurant model is given and analysed using the techniques mentioned in the paper. The benefits of using an SPN to model the restaurant system rather than using Petri net are also given.
  Keywords: Incidence matrix ; Petri net ; Signed Petri net.
  DOI: 10.1504/IJCSM.2020.10047015
   
  
• COVID-19: Machine Learning Methods Applied for Twitter Sentiment Analysis of Indians Before, During and After Lockdown  
  by H.S. Hota, Dinesh Kumar Sharma, Nilesh Verma 
  Abstract: This paper emphasizes the analyzing sentiment of Indian citizens based on Twitter data using Machine Learning (ML) based approaches. The sentiment of about 1,51,798 tweets extracted from Twitter social networking and analyzed based on tweets divided into six different segments, i.e., before lockdown, first lockdown, lockdown 2.0, lockdown 3.0, lockdown 4.0 and after lockdown (Unlock 1.0). Empirical results show that ML-based approach is efficient for Sentiment Analysis (SA) and producing better results, out of 10 ML-based models developed using N-Gram (N=1,2,3,1-2,1-3) features for SA, Linear Regression model with Tf-Idf (Term Frequency Inverse Term Frequency) and 1-3 Gram features is outperforming with 81.35% of accuracy. Comparative study of the sentiment of the above six periods indicates that negative sentiment of Indians due to COVID-19 is increasing (About 4%) during first lockdown by 4.0% and then decreasing during lockdown 2.0 (34.10%) and 3.0 (34.12%) by 2% and suddenly increased again by 4% (36%) during 4.0 and finally reached to its highest value of 38.57% during unlock 1.0.
  Keywords: Machine Learning (ML); Twitter; Sentiment Analysis (SA); Logistic Regression; COVID-19; Lockdown.
  
  
• Artificial bee colony algorithm with distance factor  
  by Min Zhou, Runxiu Wu, Hui Sun 
  Abstract: Aiming at the shortcomings of standard artificial bee colony (ABC) algorithms, such as weak local searching ability, poor diversity and easy to fall into local optimum, we propose the ABC algorithm with distance factor (DF_ABC). With the current optimal honey source as the reference, the new algorithm introduces the honey source distance reflecting the difference of the honey source location, and defines the distance factor controlling the searching direction of the algorithm through the honey source distance. The proposed searching strategy is capable of self-adaption. When the honey source distance is large, the particle can quickly jump to the peak (valley) where the global optimal point is located with the peak (valley) jumping ability; when the distance is small, the location information of the optimal honey source is used for local searching to speed up the algorithm convergence. The effectiveness of the proposed searching strategy is verified by results of the experiment on benchmark test functions. Compared with other improved algorithms, the proposed algorithm showcases the best comprehensive performance.
  Keywords: artificial bee colony (ABC) algorithm; distance factor; honey source distance; global detection; local development.
  
  
• A Survey of Blockchain : Concepts, Applications andChallenges  
  by Abhishek Taparia, Nizar Banu P K 
  Abstract: With the development of Bitcoin, organizations, be it businesses or institutions, are centring on leveraging Bitcoin's blockchain technology to non-monetary based applications to improve efficiency of the activities. Having various benefits like anonymity, decentralized, audibility etc. blockchain technology can be vastly implemented in various sectors other than financial too. This paper gives an overview the blockchain technology. It briefs about various technical concepts used in the blockchain, its types and where it can be used. It also discusses some proposed applications of the technology and tools or frameworks that can be used to develop such. It also presents the limitations of the technology.
  Keywords: Keywords: Blockchain; Hash Cryptography; Mining; Hyperledger; Proof-of-Work; Consensus Protocol; Smart Contracts.
  
  
• Half-Sweep RSOR Iteration with Three-Point Linear Rational Finite Difference Scheme for Solving First-Order Fredholm Integro-Differential Equations  
  by Ming-Ming Xu, Juamt Sulaiman, Labiyana Hanif Ali 
  Abstract: In this paper, we establish the three-point newly half-sweep linear rational finite difference-quadrature discretization scheme, which is the combination of the three-point half-sweep linear rational finite difference (3HSLRFD) scheme alone with the first-order quadrature scheme especially half-sweep composite-trapezoidal (HSCT) in discretizing the first-order linear Fredholm integro-differential equation (FIDE). Based on this established discretization scheme, the corresponding 3HSLRFD-HSCT approximation equation can be derived and then generate the large-scale and dense linear system. Furthermore, the numerical solution of the first-order linear FIDE can be obtained by implementing the Half-Sweep Refinement of Successive Over-Relaxation (HSRSOR) iterative method to solve the linear system. For the sake of comparison, the formulation of the full-sweep Gauss-Seidel (FSGS) and full-sweep Refinement of Successive Over-Relaxation (FSRSOR) methods are also presented as the control method. Finally, several numerical examples of the proposed problem are shown to demonstrate that the HSRSOR iterative approach gives the highest degree of supremacy in terms of number of iterations and execution time as compared to the other two existing methods.
  Keywords: First-order integro-differential equations; Half-sweep concept; RSOR iteration; Linear rational finite difference; Composite trapezoidal.
  
  
• Efficient projective algorithm for linear fractional programming problem based on a linear programming formulation  
  by Ahlem Bennani, Djamel Benterki 
  Abstract: In this paper, we are interested in solving a linear fractional programming problem that is converted into an equivalent linear program. The obtained problem is solved through an interior point method. In a first step, an adequate formulation of linear fractional programming problem into an equivalent linear program was proposed by Bennani et al., avoiding the increase of the dimension of the initial problem. Moreover, we successfully established a comparative numerical implementation of Ye-Lustig's algorithm to find the optimal solution. A comparative numerical study is carried out between this formulation and another classical one. The results obtained were very encouraging and showed clearly the impact of this formulation.
  Keywords: Linear fractional programming; Linear programming; Interior point method; Projective method.
  
  
• Exact reliability formula for n-clients computer network with catastrophic failure and copula repair  
  by Praveen Kumar Poonia 
  Abstract: In this paper, I have considered a general warm standby repairable k-out-of-n computer lab network with similar computers and all the computers are connected in parallel to a data server and a router. Failure rates of all n computers, data server and router are assumed to be constant and follow exponential distribution. Upon failure, every component moves into repair space and the repair supports general distribution and Goumbel-Hougard copula distribution. The objective of this paper is to evaluate the exact formulas for availability of the system, reliability of the system, mean time to failure and expected profit analysis in a way that numerical solutions can be obtained systematically in a reasonable computational time. This makes the computation uncomplicated and accurate. The problem is modelled as a finite series using supplementary variable technique, Laplace transform and copula repair. Lastly, the model is illustrated with graphs and an example for specific values of n and k.
  Keywords: k-out-of-n: G; computer network; availability; database server; catastrophic failure; Gumbel-Hougaard copula distribution.
  
  
• An Enhanced Harmony Search Integrated with Adaptive Mutation Strategy  
  by Ying Deng, Yiwen Zhong, Lijin Wang 
  Abstract: Aiming at making improvements on solutions to function optimization problems, an enhanced harmony search, called EHS, is proposed by hybridizing differential mutation strategies. EHS employs the differential mutation strategies after a solution generated by harmony search, then the solution is integrated into the differential mutation strategies as a target or current vector. Moreover, four differential mutation operators, including target-to-rand/1, target-to-rand/2,rntarget-to-best/1, and target-to-best/2, are invoked adaptively in a random way.rnExtensive experiments on CEC2014 benchmark functions demonstrate EHS is effective and efficient with the combination of harmony search and the differential mutation strategies.
  Keywords: harmony search; differential mutation; adaptive strategy; current vector; constructive algorithm.
  
  
• A ranking paired based artificial bee colony algorithm for data clustering  
  by Haiping Xu, Zhengshan Dong, Meiqin Xu, Geng Lin 
  Abstract: Data clustering aims to partition a dataset into $k$ subsets according to a prespecified similarity measure. It is NP-hard, and has lots of real applications. This paper presents a ranking paired based artificial bee colony algorithm (RPABC) to solve data clustering. First, a chaotic map is employed to generate initial food sources. Second, in order to speed up the search, RPABC uses a ranking paired learning strategy to produce new positions. Finally, the best food source is utilized to guide the search in the onlooker bees\' phase. Several datasets from the literature are used to test the RPABC. The computational results show that the proposed method is able to provide high quality clusters, and is more stable than the compared algorithms.
  Keywords: data clustering;artificial bee colony;ranking.
  
  
• Sinc Collocation Method: Solution of a Class of Strongly Nonlinear Two-Point Boundary Value Problems  
  by Mohammad Nabati, Ali Barati, Mehdi Jalalvand, Jalil Rashidinia 
  Abstract: In this study, Sinc-collocation methods based on single and double exponential transformations for finding solution of a class of nonlinear second order two-point boundary value problems were developed, and their properties were enumerated. The presented methods are shown to reduce the solution of nonlinear two-point boundary value problems to the system of nonlinear algebraic equations. The convergence and error analysis of the method has been investigated, also the upper bound of the error has been calculated as exponential form. To show the efficiency, ability and high accuracy of the method, several examples have been considered. The obtained results of Sinc methods based on single and double exponential transformations were compared with each other, and also with those of the existing numerical results of methods reported in the literature. The numerical results confirm that these methods rapidly converge and have a considerably efficient and accurate nature.
  Keywords: Sinc function; collocation method; nonlinear problems of BVPs; single and double exponential transformations.
  
  
• Method of characteristic points for composite Rydberg interatomic potential  
  by Takalani Malange, Samuel Surulere, Michael Shatalov, Andrew Mkolesia 
  Abstract: The Interpolation function in Mathematicatextsuperscript{textregistered} was used to identify the experimental data sets of copper atom as a potential energy curve. The characteristic points of the resulting energy curve were considered in three domains, each having five, four and two constraints respectively. The analytic forms of the extended-Rydberg potential (cubic, quartic and quadratic) were used for the curve fitting of the estimated parameters for the potential energy curve. The unknown parameters of each respective analytic form of the extended-Rydberg potentials were estimated using the minimization of the formulated goal function. This was done by an effective one-dimensional search for the (alpha_i)-parameter ((i=1,ldots, 3)). The results of the estimated values of the potential energy curve using the experimental data values indicate that the method of characteristic points gave modestly good estimates for the characteristic points of the composite Rydberg interatomic potential.
  Keywords: extended-Rydberg potential; composite potential; characteristic points; energy potentials; minimization.
  
  
• Source Selection and Transfer Defect Learning based Cross-Project Defect Prediction  
  by Wanzhi Wen, Ningbo Zhu, Bingqing Ye, Xikai Li, Chuyue Wang, Jiawei Chu, Yuehua Li 
  Abstract: Software defect is an important metrics to evaluate software quality. Too many defects will make the software unavailable and cause economic losses. The aim of SDP (Software Defect Prediction) is to find defects as early as possible. Based on this, source project selection and transfer defect learning based cross-project defect prediction STCPDP is proposed. This method firstly sets the threshold of the metrics to predicting the defect more effectively, secondly computes the similarity between different project versions to find the appropriate train sets, and finally combines the popular transfer defect learning method TCA+ to predict software defects based on the logistic linear regression model. Experimental results show that when the defect probability threshold is about 0.4, STCPDP has better performance based on F-measure metric, and STCPDP can effectively improve the popular CPDP models.
  Keywords: cross-project defect prediction; feature selection; logistic regression; source project selection; transfer defect learning.
  
  
• Improved rough K-means clustering algorithm based on firefly algorithm  
  by Ye Ting Yu, Jun Ye, Lei Wang 
  Abstract: The rough K-means clustering algorithm has a strong ability to deal with data with uncertain boundaries. However, this algorithm also has limitations such as sensitivity to initial data selection, as well as it use of fixed weights and thresholds, which results in unstable clustering results and decreased accuracy. In response to this problem, combined with the firefly algorithm, the original algorithm has been improved from three aspects. Firstly, based on the ratio of the number of objects in the dataset to the product of the difference of the objects in the dataset, a more reasonable method of dynamically adjusting the weights of approximation and boundary set is designed. Secondly, a method of adaptively realizing the threshold ? associated with the number of iterations is given. Then, by constructing a new objective function, and take the objective function value as the firefly brightness intensity to perform the search and update iteration of the initial cluster center point, the optimal solution obtained by each iteration of firefly is taken as the initial center position of the algorithm. Experiment result shows that the new algorithm has improved the clustering effect.
  Keywords: Rough K-means algorithm; firefly algorithm; Cluster center; Lower approximation and boundary set; Objective function.
  
  
• Deep denoiser prior and smoothed projection landweber inspired block-wise compressed sensing  
  by Chunmei Zong 
  Abstract: How to use effective image prior to reconstruct high-quality images is a key problem in compressed sensing reconstruction. By introducing instantiation priors, traditional optimization model-based compressed sensing reconstruction methods enjoy good structural analysis ability. To further improve the reconstruction quality, the optimization model-based method is combined with deep learning to introduce a deep denoiser prior into BCS-SPL algorithm via a plug and play technique. Notably?the denoising operator is obtained by training a multi-scale residual network with data-driven discriminant learning method. Multi-scale network can extract different scale feature information about the image, and the introduced deep prior is beneficial for reconstructing high-quality images. Experimental results exhibit that the proposed method can effectively improve the image reconstruction quality without the expense of too much computational complexity.
  Keywords: compressed sensing; deep learning; plug and play; deep denoiser prior.
  
  
• Machine Learning Comparative Study for Human Posture Classification using Wearable Sensors  
  by Aaron Rababaah 
  Abstract: Human posture classification plays important role in number of applications including elderly monitoring, workplace ergonomics, sleeping patterns studies, sports, fall detection, etc. Despite of the fact that the topic is well-studied in the literature, many studies utilize one to few models to investigate the classification reliability of different postures. In this paper we present a rich study of the problem with six primary machine learning algorithms and an overall of nine different models considered in training and testing the real world collected data of human subjects. In this study, six different postures are addressed namely: sleeping, sitting, standing, running, forward bending and backward bending. Two accelerometers were attached to the chest and thigh areas of human subjects where each sensor produced three different readings for x, y, and z axes. A total of six signal readings were collected per each posture which made-up the feature vector. Close to 45000 samples were recoded for all postures to be used for training and testing different machine learning algorithms. The study considered two categories of models, supervised and unsupervised learning algorithms namely: Neural Network - Multi-layer perceptron, Nearest neighbor classification, Discriminant analysis, Self-organizing maps, K-Means and Gaussian mixture model. After intensive training and testing of all algorithms, Multi-layer perceptron and K-Means outperformed other algorithms with an impressive classification accuracy of 99.88% and the lowest performing algorithm at 73.95% was the Gaussian mixture model as data may not follow Gaussian probability distribution.
  Keywords: Human posture; Wearable sensors; Machine learning; Neural Network; Multi-layer perceptron; Nearest neighbor classification; Discriminant analysis; Self-organizing maps; K-Means; Gaussian mixture model; clustering; classification; signal processing.
  
  
• Improved Total Difference Method (ITDM): A New Approach to Solving Transportation Problem Based on Modifications of Total Difference Method 1 and Integration of Total Ratio Cost Matrix  
  by Muhammad Sam'an, Yosza Dasril, Nazarudin Bin Bujang, Farikhin Farikhin 
  Abstract: In this paper, Initial Basic Feasible Solution is referred to as Initial Feasible Solution (IFS). There are two phases in solving transportation problem (TP). An IFS is determined in the first phase by using the least distribution cost, followed by calculation of the optimal solution through the modification of total difference method (TDM 1), integrated with total ratio cost matrix (TRCM) in the second phase. In some cases, it has been found that TP has equal values of the distribution least costs so that the existing methods generate two or more IFS values. The newly developed algorithm obtain the optimal solution of TP. A total of 26 numerical examples were selected from reputed journals to evaluate the performance of the newly developed algorithm. The computational performances were compared to the existing methods in the literature and the results showed that this algorithm not only solve TP with similar values optimal solution but also produce better minimal solutions than existing methods.
  Keywords: transportation problem; initial feasible solution; optimal solution;rntotal ratio cost matrix.
  
  
• Deep Learning of Human Posture Image Classification using Convolutional Neural Networks  
  by Aaron Rababaah 
  Abstract: Human posture classification plays important role in number of applications including elderly monitoring, workplace ergonomics, sleeping patterns studies, sports, fall detection, etc. In this paper a study of deep learning applied to human posture image classification using convolutional neural networks (CNNs) is presented. Typical computer vision workflow includes in the early stages: data conditioning, feature extraction, dimensionality reduction/feature selection whereas, in CNNs, these stages are not required which provides a big advantage of automatic feature extraction. In this work, CNNs are applied to human posture classification. The input data is collected using an RGB digital camera within an indoor environment targeting 10 different postures from 4 different human subjects including standing with 5 different variations, sitting with 2 different variations, bending and sleeping with two different variations. More than 6000 samples were collected for training and validation. Since number of features is among the most important parameters of CNN models, 7 independent experiments were conducted each of which has a different number of filters/kernels ranging within [1, 32]. The results of the experimental work showed that number of features influenced the classification accuracy significantly as the lowest CNN model produced 91.76% and the highest model produced 98.57% classification accuracy.
  Keywords: deep learning; convolutional neural networks; human posture classification; image processing; machine vision.
  
  
• A new artificial bee colony algorithm based on modified search strategy  
  by Kai Li, Minyang Xu, Tao Zeng, Tingyu Ye, Luqi Zhang, Wenjun Wang, Hui Wang 
  Abstract: Artificial bee colony (ABC) is an efficient global optimization algorithm. It has attracted the attention of many researchers because of its simple concept and strong exploration. However, it exhibits weak exploitation capability. To improve this case, a novel ABC with modified search strategy (namely MSABC) is proposed in this work. In MSABC, some modified elite solutions are preserved and used to guide the search. In addition, MSABC uses the modified elite solutions to generate offspring to replace the probability selection in the onlooker bee phase. To evaluate the capability of MSABC, 22 classical problems are tested. Results demonstrate MSABC achieves superior performance than five other ABC variants.
  Keywords: Artificial bee colony (ABC); Elite solution; Search strategy; Probability selection.
  
  
• A method of designing swinging-leg walking trajectory for biped robot on plat ground  
  by Yingli Shu, Quande Yuan, Jian Zhang, Huazhong Li, Yuzhen Pi, Wende Ke 
  Abstract: The periodic walking of biped robot involves the alternate movement of supporting leg and swinging leg. In order to quickly plan the gait, it is necessary to select the key posture of biped walking on the premise of maintaining the stability of the robot. Based on the known information, the spline curve is designed and solved to construct the ankle trajectory of the swinging leg of the robot. Simulation results showed the feasibility of the method.
  Keywords: biped robot; trajectory; walking; zero moment point (ZMP).
  
  
• Safety and Energy-saving Driving Behavior Evaluation with Driving Feature Constraint TOPSIS Method  
  by Xinlei Wei, Yingji Liu, Wei Zhou, Haiying Xia, Xuan Dong 
  Abstract: With the development of road traffic, transportation enterprises pay more attention to safety and environmental protection, and the safety and energy-saving evaluations are of great significance in the management of vehicles. In view of the different characteristics of the evaluation system in the TOPSIS evaluation method, the objectivity and accuracy of the evaluation are reduced. This method takes into account the characteristics of the bad driving behavior index system based on the GPS data and improves the TOPSIS evaluation method by using index extremum constraint on the evaluation process. The experimental results show that the proposed method improves the objectivity and accuracy of the evaluation, which is always the same as the actual situation.
  Keywords: Safety and energy-saving evaluation; TOPSIS; Bad driving behavior.
  
  
• KERNEL BASED APPROXIMATION OF VARIABLE-ORDER DIFFUSION MODELS  
  by Marjan Uddin, Muhammad Awais 
  Abstract: In this paper, a numerical scheme is constructed which is based on radial basis functions (RBF) and Coimbra variable time fractional derivative of order 0 < A(t,x) < 1. The derivative due to Coimbra can efficiently modelrna dynamical system whose fractional order behavior varies with time as wellrnas space. The stability, convergence of the RBF based numerical scheme isrndiscussed and the developed numerical scheme is validated for various 1D andrn2D anomalous diffusion models with different fractional variable order eitherrna function of t or x . The accuracy and efficiency of the numerical scheme isrnachieved by comparing the results for available results in the literature.
  Keywords: RBF; Variable order; Fractional order; Anomalous Diffusion; Numerical approximation; Coimbra Derivative.
  
  
• IRPSM-Net:Information Retention Pyramid Stereo Matching Network  
  by Yun Zhao, Jiahui Tang, Xing Xu, Xiang Zhou 
  Abstract: In order to prevent the lack of information in the stereo matching process and improve the disparity map accuracy. The information retention pyramid stereo matching network (IRPSM-Net) was proposed a novel architecture that can relieve the limitation of accuracy and retention the original information of the image. The proposed network consisted an information retention pyramid module (IRPM) without batch normalization to retain the image information. And the training process was optimized by group normalization, which further improves the effect of stereo matching. The ablation experiments show that our method can effectively improve the accuracy of 0.17% in the threshold 3 pixels of KITTI2012 stereo dataset and 0.09% in the whole region of KITTI2015 stereo dataset. It showed that the improvement of IRPSM-Net can effectively improve the quality of the generated disparity map.
  Keywords: Stereo matching;Multi-scale;Information retention pyramid;Group normalization.
  
  
• An enhanced multi-objective particle swarm optimization with levy flight  
  by Hai-ying Lan, Gang Xu, Yu-qun Yang 
  Abstract: In the scope of multi-objective particle swarm optimization (MOPSO) research, avoiding premature convergence remains a challenge. To address this issue, the article develops an enhanced multi-objective particle swarm optimization with Levy flight (LF-MOPSO). In LF-MOPSO, swarm is made to evolve based on the original MOPSO to accelerate convergence. Then, Levy flight is adaptively activated to maintain diversity, so as to deal with the premature convergence when Pareto frontier is stagnant. It realizes the transformation between shrinkage and divergence of population diversity by self-adaptive conversion mechanism, which further improves the search ability of MOPSO. LF-MOPSO has been contrasted with some recently improved MOPSOs, the experimental outcomes indicate that LF-MOPSO ensures the better approximation to the Pareto optimal frontier, and gains the non-dominated solutions with good diversity and distribution.
  Keywords: Multi-objective optimization; Particle swarm optimization; Levy flight; Non-dominated solution.
  
  
• VGBNet: A Disease Diagnosis Model Based on Local and Global Information Fusion  
  by Yong LI, Xinyu ZHAO, Manfu MA, Qiang ZHANG, Hai JIA, Xia Wang 
  Abstract: There are significant differences in the data volume of different types of diseases in the electronic medical record data. Moreover, mainstream auxiliary diagnosis and prediction models either ignore local information or ignore global information. In response to these problems, this paper use a method of fusion random resampling to balance the data set, Using graph convolutional neural network to extract global features, combined with a bidirectional self-attention network, a VGBNet model is used to link local and global features to achieve diagnosis and prediction of diseases. This model can not only deal with unbalanced data but also combine global and local features to improve the accuracy of disease-assisted diagnosis and prediction. A large number of experiments show that the performance of this model has improved compared with BERT and GCN. This is of great significance to the precise auxiliary diagnosis of diseases.
  Keywords: Unbalanced Data Set; Disease Prediction; Graph Convolutional Neural Network; Attention Mechanism.
  
  
• Dense Deep Stochastic Configuration Network with Hybrid Training Mechanism  
  by Weidong Zou, Yuanqing Xia, Weipeng Cao 
  Abstract: Thanks to the supervised parameter generation strategy and non-iterative training mechanism, Deep Stochastic Configuration Network (DSCN) has achieved very efficient modeling efficiency in scenarios with relatively small problem complexity. However, the increasing number of hidden layers and the amount of training data have issued a challenge to the implementation of DSCN. To solve this problem, we propose a Dense DSCN with a Hybrid Training mechanism (HT-DDSCN), which extends the network structure of the DSCN to a dense connection type and combines three typical optimization techniques and one universal control strategy to optimize the calculation process of the output weights. Extensive experiments on four benchmark regression problems show that HT-DDSCN can significantly improve the generalization ability and the stability of DSCN.
  Keywords: Deep Stochastic Configuration Network; Randomized Neural Networks; Generalization Ability.
  
  
• Time series granulation-based multivariate modeling and prediction  
  by Mengjun Wan, Hongyue Guo, Lidong Wang 
  Abstract: The typical characteristics of time series data exhibit a large data size, high dimensionality, and high correlation. To better extract high-level representative information for time series, this study proposes a novel granular vector autoregressive (GVAR) model, which incorporates granular computing with VAR models to predict the main varying ranges of the multivariate time series. The proposed model first utilizes the principle of justifiable granularity to construct information granules, which capture the cardinal information hidden in the time series. Then, the granular VAR model is built based on the upper and lower bounds of the constructed information granules simultaneously. Here, the interval least squares (ILS) algorithm is employed to estimate the models coefficients, and the regressive order is determined by the Bayesian information criterion (BIC). Finally, experimental studies are conducted to illustrate the effectiveness and practicality of the proposed prediction model.
  Keywords: Information Granule; Multivariate Time Series; Granular Prediction.
  
  
• Theoretical Model Analysis and Case Simulation of Spherical Involute Surface  
  by Zixia Chen, Yan Zhao, Jingyu Liu, Zelin Chen 
  Abstract: The theoretical model of spherical involute surface is the premise and foundation of 3D modelling design technology of spiral bevel gear. The forming process of the surface is also the forming process of its mathematical model. Based on the generating principle of spherical involute on cone and surface analysis, the system model is gradually divided into blocks from curve to surface, and the surface of bevel gear is obtained by parameter coupling. The modelling process of curve, surface or initial spiral can be independently referenced. It can realize the precise modelling of complex spiral profile surfaces such as variable spiral angles. In this paper, the mathematical model of spiral bevel gear was simulated and analyzed on the MATLAB2019b platform. The simulation results had verified the correctness of the algorithm of the spiral bevel gear model and the practicability of design optimization.
  Keywords: spiral bevel gear; spherical involute surface; mathematical modelling; MATLAB simulation.
  
  
• Multi-objective Cellular Memetic Algorithm  
  by Xianghong Lin, Tingyu Ren, Jie Yang, Xiangwen Wang 
  Abstract: This paper presents a multi-objective cellular memetic algorithm (denoted by MOCMA) based on k-means clustering, which integrates the clustering-based local search method into multi-objective cellular genetic algorithm. Specifically, according to the objective function values of individuals in each generation, the k-means clustering is used to control the similar individuals gathered in a cluster. Meanwhile, to explore the search space efficiently and get the Pareto optimal solutions in objective space, one individual is selected randomly to undergo local search from each cluster and it will be improved than before. The MOCMA is applied to constrained and unconstrained problems. We analyze the influence of cluster number on the performance of the algorithm, and compare the MOCMA with other evolutionary multi-objective optimizers. It indicates that the proposed MOCMA is efficient for solving the multi-objective optimization problems.
  Keywords: cellular genetic algorithm; multi-objective optimization; k-means clustering; pareto optimal set; local search.
  
  
• Research on x-vector speaker recognition algorithm based on Kaldi  
  by Hong Zhao, Lupeng Yue, Weijie Wang, Xiangyan Zeng 
  Abstract: This paper presents a convolutional neural network with an attention mechanism for analyzing the spectrogram in an x-vector based speaker recognition system. First, the convolutional neural network (CNN) is used to extract the features of the spectrogram. Then, an attention mechanism is designed to calculate the frame weight in the statistical pooling layer. Finally, probability linear discriminant analysis (PLDA) is used as a back end classifier. The system is implemented using Kaldi speech recognition tools and tests on the Voxceleb1 database. The experimental results show that the combination of spectrogram and CNN gains a relative improvement of 6.7% in equal error rate (EER) compared with the x-vector baseline system. The attention mechanism for the statistical layer further leads to a relative improvement of 26.1%. Overall the proposed method outperforms state-of-the-art methods on the Voxceleb1 database.
  Keywords: spectrogram; attention mechanism; x-vector; speaker recognition; Kaldi.
  
  
• Framework and Experimental Analysis of Generalized Surrogate-Assisted Particle Swarm Optimization  
  by Rui Dai, Jing Jie, Hui Zheng, Miao Zhang, Yixiao Lu 
  Abstract: The paper develops a framework of generalized surrogate-assisted particle swarm optimization (GS-PSO) to solve computationally expensive problems. To ensure the generalization ability and optimization accuracy of the algorithm, some researches about the factors of GS-PSO and selection of surrogates have been done. Some statistics indexes such as accuracy, robustness, and scalability are formulated to evaluate six popular metamodels, which can help to choose the proper surrogates for GS-PSO during the optimization process. A series of simulation experiments are conducted based on some notable benchmark functions. The results show that GS-PSO with RBF is a robust surrogate-assisted algorithm for computationally expensive problems. Meanwhile, a proper combination of optimizers and surrogates can contribute to an improvement of GS-PSO for different optimization problems.
  Keywords: SAEAs; Surrogate models; Optimization algorithms; Performance indexes.
  
  
• Underwater Short Distance Magnetic Communication Based on Coupling Coils in Sealed Metal Bin  
  by Qi Chen, Linqi Xia, Wei Zhang 
  Abstract: With the more and more extensive demand of ocean exploration, the communication system of deep-water operation equipment in the ocean requires higher and higher performance. Due to the absorption of electromagnetic wave by seawater, underwater wireless communication is very difficult. In order to adapt to the complex working environment in the deep sea, it is necessary to adopt the communication system with the characteristics of non-contact, low power consumption, small size, two-way transmission and so on. In this paper, a set of underwater short distance magnetic coupling communication system equipment is developed. Combined with the parameter calculation formula of magnetic coupling coil, the short spiral coil is wound, and the experiment is carried out to simulate the seawater environment. The experimental results show that the system has the advantages of low power consumption, small volume, two-way transmission and good stability.
  Keywords: underwater; magnetic coupling; Non contact.
  
  
• A Real-coded Chicken Swarm Optimization algorithm for solving Traveling Salesman Problem  
  by Min Lin, Yuhang Yang, Yiwen Zhong, Juan Lin 
  Abstract: Chicken swarm optimization (CSO) algorithm, which is inspired by the hierarchal structure and the behaviours of the chicken flock, was first presented for continuous optimization problems. The paper proposes a real-coded scheme of CSO algorithm (RCCSO) to solve Traveling Salesman Problem (TSP). In the RCCSO algorithm, each position vector represents a visiting sequence of cities. In a position vector, each dimension represents a city and is coded with a real number. The integer part of the real number represents the index number of the city, and the decimal part denotes the visiting order of the city. Using this coding scheme, the discrete neighbourhood of TSP is converted into a continuous neighbourhood. Two repair operators, relocation operator and replacement operator, are designed to guarantee that position vector is always a valid solution of TSP. Finally, the RCCSO algorithm is compared with many different types of intelligent optimization algo-rithms. Experimental results prove that the RCCSO algorithm can find the shortest path more quickly and effectively on most TSP data sets.
  Keywords: Chicken Swarm Optimization; Real-coded scheme; Traveling Salesman Problem; Swarm intelligence algorithm; Relocation operator; Replacement operator.
  
  
• Enhanced Image Super-Resolution Using Hierarchical Generative Adversarial Network  
  by Jianwei Zhao, Chenyun Fang, Zhenghua Zhou 
  Abstract: Recently, generative adversarial networks (GAN) have been introduced in single-image super-resolution (SISR) to reconstruct more realistic high-resolution (HR) images. In this paper, we propose an effective SISR method, named super-resolution using hierarchical generative adversarial network (SRHGAN), based on the idea of GAN and the prior knowledge. Different from the existing GANs that focus on the depth of networks, our proposed method considers the prior knowledge in addition. That is, we introduce an edge extraction branch and an edge enhancement branch into GAN for considering the edge information. By means of the added edge loss in the loss function, the edge extraction branch and the edge enhancement branch will be trained to reconstruct the sharp edge well. Experimental results on several datasets illustrate that our reconstructed visual effect images are clearer and sharper than some related SISR methods.
  Keywords: Image Super-Resolution; Deep Learning; Generative Adversarial Network; Hierarchical Network.
  
  
• Learning-assisted intelligent risk assessment of highway project investment  
  by Hongwei Liu, Zihao Zhang 
  Abstract: Highway project has the characteristics of long construction period, large investment scale, high investment risk and high technical requirements. How to identify, classify and evaluate the project investment risk is one of the hot spots of domestic and foreign experts. Aiming at the problem of investment risk management, this paper takes 15 highway investment projects in the past decade as the research object. According to the problems found in the actual process of highway project development, the investment risk classification is realized, and the investment risk index system including 12 first-level indexes and 30 second-level indexes is established. The hierarchical weight model of highway engineering investment risk assessment is proposed. The intelligent evaluation of highway engineering investment risk by extreme learning machine and broad learning system algorithm is discussed. The comparative experimental results show that the improved intelligent evaluation model can evaluate and predict the investment risk of highway engineering projects more effectively. The R-square value of the improved intelligent evaluation model is increased by 0.35, and the accuracy is greatly improved. It can provide decision support for highway engineering project investment risk management.
  Keywords: risk assessment; highway; risk index system; extreme learning machine; broad learning system.
  
  
• Research on the Construction of E-commerce Valley Industry-Education Integration Platform  
  by Juyan Che, Jiangjun Yuan, Jie Wang, Jiawen Shi, Weinan Liu 
  Abstract: In response to the nation strategies of the one hand initiative and the Belt and Road Initiative, Hangzhou's development strategy of building an international e-commerce center has been advanced. The "E-commerce Valley" project is a smart building system with a variety number of applications of IoT. With the development of IoT application, WSN develops quickly. WSN is usually deployed in outdoor space, which makes the security of network transmission more important. In this paper, we discuss how to reduce the computing efficiency of aggregators in data aggregation applications. The traditional way to implement this method is to use the for-loop to process. The computational performance of this method has not lived up to expectations. We use the built-in sum function instead of the for-loop to implement the protocol. Through two experiments, we conclude that the implementation scheme using sum function has better computational performance.
  Keywords: E-commerce Valley; Urban Logistics; Data Aggregation; Python implementation.
  
  
• Texture feature extraction method of multi-focus image based on gradient fusion rule  
  by Dr. Ding, Guotao Zhao 
  Abstract: In order to improve the texture detection ability of multi-focus images, a texture feature extraction method based on gradient fusion rule is proposed. The texture information detection and imaging model of multi-focus image is constructed. The texture block feature matching model of multi-focus image is constructed by using the method of region information block detection. Combined with gradient fusion rule model, the analytic rule function of texture information gradient fusion of multi-focus images is constructed. Through gradient fusion rule distribution and texture information feature decomposition results of images, multi-order moment feature decomposition is adopted to extract texture features of multi-focus images. The simulation results show that the texture feature extraction of multi-focus images by this method has higher output recognition and better texture distribution fusion performance, which improves the imaging quality and texture information enhancement effect of multi-focus images.
  Keywords: gradient fusion rules; Multi-focus image; Texture; feature extraction.
  
  
• Modelling and optimal control analysis of coffee berry disease with cost-effectiveness in the presence of temperature variability.  
  by Abdisa Melese, Oluwole Makinde, Legesse Obsu 
  Abstract: In this paper, we propose and analyze a nonlinear deterministic mathematical model for the impact of temperature variability on coffee berry disease in coffee plants. In the analysis of the model, we derived the basic reproduction numbers at minimum temperature T0 and maximum temperature Tm which help us in establishing the local and global stability of disease-free and endemic equilibrium points. The global stability of endemic equilibrium is determined by using a Lyapunov function. We proved that the model exhibits forward and backward bifurcation by the concept of center manifold theory. Sensitivity indices are also discussed. We extend the proposed model into the optimal control problem by incorporating three controls. We also analyze the necessary conditions for the optimal control of the disease by applying Pontryagin minimum principle. In addition, we investigate the cost-effective analysis to determine the most effective strategy with minimum costs. Finally, we present the numerical simulations.
  Keywords: mathematical model; coffee berry; optimal control; cost-effectiveness; temperature variability.
  
  
• HYBRID MODEL FOR CLASSIFICATION OF DISEASE USING DATA MINING WITH PARTICLE SWARM OPTIMIZATION TECHNIQUE  
  by Akhilesh Shrivas, Rashmi Gupta, Ragni Shukla 
  Abstract: This paper presents the hybrid model based on Particle Swarm Optimization (PSO), K-means, and Self Organizing Map (SOM), combined with different data mining-based classifiers for identification and classification of various health-related diseases. The main contribution of this research work is to develop a robust and computationally efficient predictive hybrid model using K-means and SOM unsupervised clustering techniques to facilitate the classification of data. The clustering algorithms help to reduce the unclustered instances or wrongly instances from the database, while the PSO is used to optimize the features of datasets. Both of these methods help to improve classification accuracy and reduce uncertainty. The proposed hybrid-based model diagnosed different diseases, namely Chronic Kidney Disease (CKD), Breast cancer, and Hepatitis disease, is one of the best results compared with other results reported after comparison in this literature. The results confirmed that the proposed hybrid system achieved better performance in measures of accuracy, sensitivity, and specificity.
  Keywords: hybrid model; classification; data mining; self-organizing map; SOM; particle swarm optimization; PSO; decision tree; DT.
  
  
• A personalized travel route recommendation method based on improved greedy algorithm  
  by Qun Shang 
  Abstract: In order to improve the low accuracy, recall rate and popularity of the traditional method, this paper proposes a personalized tourism route recommendation method based on the improved greedy algorithm. Firstly, the attractiveness rating index of tourist attractions is established, the weight of the index is determined by the normalization method, and the attractiveness rating is obtained by combining the condition evaluation matrix and potential evaluation matrix. Then, the greedy algorithm is improved by decomsolving the maximal link subgraph to find the biggest node of influence, and the individual contribution value of tourism route resource knowledge is calculated by using the improved greedy algorithm. Finally, personalized tourism route resources are recommended according to different individual contribution values. The results show that the maximum recommendation accuracy of this method can reach 96%, the maximum recall rate can reach 80%, and the maximum is close to 4.8.
  Keywords: Maximal unicom subgraph; Perceptual sorting; Greedy decision-making range; Attractiveness rating; Improved greedy algorithm; Personalized recommendation.
  
  
• Improved Handwritten Digit Recognition Using Artificial Neural Networks  
  by Debabrata Swain, Badal Parmar, Hansal Shah, Aditya Gandhi 
  Abstract: Handwritten digit recognition is one of the major challenging problems concerning real-time applications. It has found utilization in various fields like postal mail arranging and healthcare. There is a necessity for a framework that can apprehend the penmanship of all age groups with precision. In our proposed system, an acute number recognition system is implemented by neural networks. Different types of optimizers are employed to enhance the performance of the system. An optimizer is an essential part that aids with tracking down the ideal arrangement of weights and their values for improving accuracy. The Modified National Institute of Standards and Technology (MNIST) digit dataset has been utilized for experimentation. This work addresses how the general accuracy of a neural network can be improved by utilizing more befitting optimizers. Extensive experimentation was performed, and the model achieved recognition accuracy of 99.87% with an RMSProp optimizer.
  Keywords: handwritten digit recognition artificial neural networks; optimizers; MNIST dataset; hyper-parameters; deep learning; Adagrad; Adam; RMSprop; Gradient Descent; Stochastic gradient descent; Mini-Batch gradient descent.
  
  
• Solving First-Order Fuzzy Initial Value Problems Using One-Step Scheme with Second and Third Fuzzy Derivatives  
  by Kashif Hussain, Oluwaseun Adeyeye, Nazihah Ahmad 
  Abstract: Fuzzy differential equations model the uncertain behaviour of the dynamic model. In cases where the exact solution does not exist, numerical methods are adopted to obtain an approximate solution of these equations. Although there are several numerical approaches in the literature, the accuracy of the existing numerical methods for solving linear and non-linear first-order fuzzy initial value problems (FIVPs) in terms of absolute error could be improved. For this reason, this article develops a one-step implicit method with the presence of higher fuzzy derivatives to obtain the numerical solution of both linear and non-linear first-order FIVPs. First, the convergence properties of the one-step method are described in detail using the definition of zero-stability and consistency for linear multistep methods. Then the one-step scheme with second and third fuzzy derivatives is adopted to solve some first-order FIVPs. The results indicate that the proposed method efficiently solves fuzzy initial value problems.
  Keywords: Fuzzy Set Theory; Fuzzy Derivatives; Fuzzy Initial Value Problems; First-Order; One-step; Implicit Method.
  
  
• Modularization interface mechanism supported by equilibrium measurement of cloud logistics platform for smart communities  
  by Yuyan Shen 
  Abstract: It has put forward the concept of the smart community in Zhejiang Province of China. The new retail business form of online and offline + smart logistics has broken the service boundary of traditional logistics function system and put forward new requirements for modular cloud logistics service platform. This paper is intended to analyze especially the types of interfaces (ie, module the design rules of service modularization. In order to achieve the optimization, the logistics equilibrium measurement model of cloud logistics platform is to delivered. It is constructed on the proposed concept of logistics equilibrium based on the theory of logistics equilibrium and the improvement rate of workflow. In the construction of the model, the four elements of time, space, quantity, and structure in the logistics equilibrium theory are set as equilibrium constraints. To obtain the competitive advantage of the logistics system, the relative importance of the flow volume and flow rate corresponding to the equilibrium elements is be analyzed, with the improvement rates of flow volume and flow rate as the main factors affecting equilibrium. In addition, the specific application of equilibrium theory is illustrated by 56 Tongcheng examples of cloud logistics platforms to improve and enrich the logistics equilibrium theory of cloud logistics platforms, which proves to enhance capabilities, and improve the competitiveness of cloud logistics service platforms. It expands the application scenarios of logistics equilibrium theory, uses smart data technology, integrates the functional modules of community express and retail distribution, and creates a cloud logistics service system of smart community in future.
  Keywords: Cloud Logistics; Modularization; Interface Mechanism; Equilibrium Measurement?“smart community”.
  
  
• Trajectory Tracking Based on Adaptive Fast Non-singular Terminal Sliding Mode Control  
  by Jiqing Chen, Xu Liu, Chaoyang Zhao, Rongxian Mo, Chunlin Huang, Ganwei Cai 
  Abstract: Aiming at the binocular vision system with dynamic errors and external disturbances, a trajectory tracking control method based on adaptive fast non-singular terminal sliding mode is proposed. Firstly, the FNSTMC is adopted to ensure that the system converges in a finite time while avoiding singularity; Secondly, the adaptive RBF neural network is used to approximate the dynamics errors, friction and external disturbance to eliminate the influence of uncertain factors; Thirdly, the PSO is used to optimize the parameters of the adaptive RBF neural network, so as to avoid the mapping failure caused by unreasonable parameter setting; Finally, the Lyapunov function is established according to the Lyapunov stability, and the theoretical stability of the system is proved. The experimental results show that the proposed trajectory tracking control method can effectively improve the convergence speed and tracking accuracy of the system, enhance the robustness and weaken the chattering of the system.
  Keywords: trajectory tracking; terminal sliding mode control; radial basis function neural network; Particle swarm optimization algorithm.
  
  
• Comprehensive assessment and prediction of urban resilience: A case study of China  
  by Yao Wang, Zhe Liu 
  Abstract: Urban resilience is widely used to describe the capability of cities to fend off internal and external risks, reduce losses and recover quickly. The assessment and prediction of urban resilience can help cities to develop strategies and plans to deal with unknown disaster risks. This paper employs the method of combining The Entropy Weights Method and Technology for Order Preference by Similarity to an Ideal Solution to establish the urban resilience comprehensive evaluation model, and uses the grey prediction model and Back propagation neural network to predict the urban disaster resilience value. The results have suggested that the resilience of cities in different dimensions is generally high in eastern China, while the urban resilience in Northeast China is not higher than the average, and the regions with an average level of resilience are concentrated in Central China.
  Keywords: Urban Resilience; EWM; TOPSIS; BP neural network; Grey model.
  
  
• Bagging Method-Based Classifier Chains for Multi-label Classification  
  by Jiaman Ding, Haotian Tan, Runxin Li, Yuanyuan Wang, Lianyin Jia 
  Abstract: Classifier chains is one of the main methods to deal with multi-label classification. For classifier chains, it is important to order labels according to the correlations among labels and to find the corresponding feature subsets. Therefore, we propose a bagging method-based classifier chains for multi-label (BMCC4ML in short). Firstly, unique feature subset for each label is selected by measuring feature importance of labels based on ReliefF. Next, considering the positive and negative correlations from a global perspective, all the labels are in order by calculating the correlations among labels. After that, inspired by ensemble learning methods, the ensemble classifier chains were constructed according to the strategy that the number of chains and the repeat rate of label order are higher than given thresholds to improve the prediction accuracy. Experiments on publicly accessible multi-label datasets demonstrate that BMCC4ML achieves more prominent results than other related approaches across various evaluation metrics.
  Keywords: Multi-label classification; Ensemble chains; ReliefF; Feature selection; Bagging Method.
  
  
• Design Of Emergency Rescue Plans And Rescue Equipment For Confined Spaces  
  by JianYe Cui, Youchun Li, Haiping Li, Shan Gao, Yue Zhou 
  Abstract: AbstractThe emergency rescue work in a confined space is not only difficult, but also has a high risk of danger. According to the pothole and small space operating environment, a pothole and small space emergency rescue program and design of the mechanical structure of the relevant equipment, the equipment through the lifting way into the pothole or small space, and approach the unconscious trapped persons, telescopic carriage to hold up part of the human body, and then use the arm tied to the wrist or ankle, after the walking mechanism will rescue objects transported to the hole, relying on lifting equipment to equipment Hang out with the rescue object.
  Keywords: Keywords: confined space; emergency rescue; rescue equipment; structural design.
  
  
• An option pricing model with adaptive interval-valued fuzzy numbers  
  by Qiansheng Zhang, Jingfa Liu, Haixiang Yao 
  Abstract: This paper proposes an option pricing model with interval-valued fuzzy interest rate, volatility and stock price. The interval-valued fuzzy pattern of Black-Scholes option formula is also investigated. With the presented option pricing model the European option price can be evaluated by adaptive interval-valued fuzzy number. Utilizing the proposed interval-valued fuzzy option valuation formula, the option investor can pick European option price with an acceptable interval belief degree for the later transaction.
  Keywords: Interval-valued fuzzy number; Option pricing; Possibilistic mean; Black-Scholes formula.
  
  
• Artificial bee colony with multiple search strategies and a new updating mechanism  
  by Xin Li, Kai Li, Tao Zeng, Tingyu Ye, Luqi Zhang, Hui Wang 
  Abstract: The imbalance of exploration and exploitation is a weakness in artificial bee colony (ABC) algorithm. To overcome this deficiency, this paper presents an improved ABC (namely IABC) by employing multiple search strategies and a novel updating method. Firstly, a concept of marginal group is introduced to construct an exploration search strategy. Then, an exploitation search strategy is designed utilizing some excellent solutions. Thirdly, the probabilistic selection strategy is modified on this basis of some elite solutions. In the experiments, 22 benchmark problems were utilized to prove the effectiveness of IABC. Test results indicate that IABC achieves stronger optimisation capabilities than the other five ABCs.
  Keywords: Artificial bee colony; marginal group; multiple search strategies; selection mechanism.
  DOI: 10.1504/IJCSM.2022.10048859
   
  
• Asymmetry approach to study for chemotherapy treatment and devices failure time's data using modified power function distribution with some modified estimators  
  by Azam Zaka, Ahmad Saeed Akhter, Riffat Jabeen 
  Abstract: In order to improve the already existing models that are used extensively in bio sciences and applied sciences research, a new class of Weighted Power function distribution (WPFD) has been proposed with its various properties and different modifications to be more applicable in real life. We have provided the mathematical derivations for the new distribution. The aim of the study is to increase the application of the Power function distribution. The main feature of the proposed distribution is that there is no induction of parameters as compared to the other generalisation of the distributions, which are complex, having many parameters. We have used R programming to estimate the parameters of the new class of WPFD using Maximum Likelihood Method (MLM), Percentile Estimators (PE) and their modified estimators. After analysing the data, we conclude that the proposed model WPFD performs better in the data sets while compared to different competitor models.
  Keywords: adequacy model; characterisation; modified estimator; percentile estimator; PFD; power function distribution; weighted distribution.
  DOI: 10.1504/IJCSM.2022.10048625
   
  
• Parameter estimation for mean-reversion type stochastic differential equations from discrete observations  
  by Chao Wei 
  Abstract: This paper is concerned with the parameter estimation problem for mean-reversion type stochastic differential equations from discrete observations. The Girsanov transformation is used to simplify the equation because of the expression of the drift coefficient. The approximate likelihood function is given, the consistency of the estimator and asymptotic normality of the error of estimation are proved. An example is provided to verify the results.
  Keywords: parameter estimation; discrete observation; consistency; asymptotic normality; mean-reversion type; stochastic differential equations; Girsanov transformation; approximate likelihood function; Burkholder-Davis-Gundy inequality; Holder inequality.
  DOI: 10.1504/IJCSM.2022.10048627
   
  
• A closed-form general solution for the distance of point-to-parabola in two dimensions  
  by Chang-Chien Chou 
  Abstract: This paper resolves a closed-form general solution for the distance from a point to a parabola in a two-dimensional plane. This is the quickest way to answer the question with exact value. Although the method for resolving the general solution is somewhat fundamental, so far there is no efficient way provided in the literature than we do in this work. The closed-form general solution takes O(1) instant time answering the question and thus should be archived. Versatile applications need to compute the distance among parabolas in 2D. Practitioners in all fields may access the source code in the appendix for rapid conducting to their applications.
  Keywords: shortest path; shortest path of parabola; computer-aided design and manufacturing; computer graphics.
  DOI: 10.1504/IJCSM.2022.10048628
   
  
• Scene text detection method research based on maximally stable extremal regions  
  by Lei Xu, Yi Liu, Lianming Mou 
  Abstract: Text information is an important basis for people to understand the natural scene image. At first, an edge-enhanced maximally stable extremal regions (MSER) text detection method based on weighted guided filtering and histograms of oriented gradients (HOG) features is proposed. Then, a two layers candidate text validation method from coarse to fine is proposed. In the first layer, a heuristic rule for validating candidate character regions is designed based on the shape features of text regions. In the second layer, the recognition of character regions is realised by using support vector machines (SVMs) with 9-dimensional features such as Hu moment invariants and stroke width transformation. The proposed method is validated by the benchmark datasets ICDAR 2013. The experimental results show that the method is comparable with other most advanced methods.
  Keywords: text detection; MSER; maximally stable extremal regions; edge enhancement; SVM; support vector machine; HOG; histograms of oriented gradients.
  DOI: 10.1504/IJCSM.2022.10048626
   
  
• An improved pseudospectral approximation of coupled nonlinear partial differential equations  
  by A.K. Mittal, L.K. Balyan 
  Abstract: In this paper, we propose a time-space Chebyshev pseudo-spectral method for the numerical solutions of coupled Burger's equation, Whitham-Broer Kaup shallow water model and coupled nonlinear reaction-diffusion equations. This technique is based on orthogonal Chebyshev polynomials which are discretised at Chebyshev-Gauss-Lobbato CGL points. A mapping is used to transform the non-homogeneous initial-boundary values to homogeneous initial-boundary values. By applying the proposed method in both time and space, the problem is reduced into a system of a nonlinear coupled algebraic equations which are solved using the Newton-Raphson method. Also present the error estimates in L₂− norm. The results obtained by the scheme are very accurate and effective. Presented numerical results confirm the spectral accuracy.
  Keywords: Coupled Burger's equation; Whitham-Broer Kaup shallow water model; coupled nonlinear reaction diffusion equations; pseudospectral method; Chebyshev-Gauss-Lobbato points.
  DOI: 10.1504/IJCSM.2022.10048842
   
  
• Scale parameter recognition of blurred moving image based on edge combination algorithm  
  by Dongbo Lv 
  Abstract: In order to overcome the problems of the traditional fuzzy motion image scale parameter recognition method with high recognition error and poor denoising effect, the paper proposes a new fuzzy motion image scale parameter recognition method based on edge combination algorithm. A fuzzy moving image denoising method based on fuzzy wavelet threshold denoising is used to remove noise in the blurred moving image. The scale parameter of the blurred moving image after denoising is extracted through the blurred moving image degradation model. According to the scale parameters of the blurred moving image, the edge combination algorithm is used to realise the recognition of the scale parameters of the blurred moving image. Experimental results show that the proposed method has the best quality of the blurred motion image after denoising, the lowest misrecognition rate of the scale parameter of the blurred motion image, and the application performance is remarkable.
  Keywords: edge combination algorithm; blurred; moving; image; scale; parameter recognition.
  DOI: 10.1504/IJCSM.2022.10048629
   
  
• To solve multi-class pattern classification problems by grid neural network  
  by Ajendra Kumar, Preet Pal Singh, Dipa Sharma, Pawan Joshi 
  Abstract: Grid computing is employed to unravel massive computational problems by using large numbers of heterogeneous computers connected to the computing network. Job scheduling is an important part of the grid computing environment, which is employed to extend the throughput and reduce the turnaround and reaction time. This paper proposed a new scheduling algorithm called "Feed forward neural network in the grid computing (FFNNGC) system," which is used to solve some real-life problems related to the pattern classification. In the proposed method, we have used a feed-forward algorithm to find the output in the grid computing network, and the network training is done until the system converges to a minimum error solution. The pattern classification problem consists of 13 real-life, and artificial dataset problems, including two class and multiclass problems. Experiments were performed under these real-life problems, and the results indicated that the proposed method is helpful in such types of problems.
  Keywords: grid computing; pattern classification; artificial neural network; distributed heterogeneous systems; feed forward algorithm; back propagation algorithm; FFNN; feed forward neural networks.
  DOI: 10.1504/IJCSM.2022.10048631