Research on Node Classification Based on Joint Weighted Node Vectors

Node of network has lots of information, such as topology, text and label information. Therefore, node classification is an open issue. Recently, one vector of node is directly connected at the end of another vector. However, this method actually obtains the performance by extending dimensions and considering that the text and structural information are one-to-one, which is obviously unreasonable. Regarding this issue, a method by weighting vectors is proposed in this paper. Three methods, negative logarithm, modulus and sigmoid function are used to weight-trained vectors, then recombine the weighted vectors and put them into the SVM classifier for evaluation output. By comparing three different weighting methods, the results showed that using negative logarithm weighting achieved better results than the other two using modulus and sigmoid function weighting, and was superior to directly concatenating vectors in the same dimension.

Photovoltaic Models Parameters Estimation Based on Weighted Mean of Vectors

Renewable energy sources are gaining popularity,particularly photovoltaic energy as a clean energy source.This is evident in the advancement of scientific research aimed at improving solar cell performance.Due to the non-linear nature of the photovoltaic cell,modeling solar cells and extracting their parameters is one of the most important challenges in this discipline.As a result,the use of optimization algorithms to solve this problem is expanding and evolving at a rapid rate.In this paper,a weIghted meaN oF vectOrs algorithm(INFO)that calculates the weighted mean for a set of vectors in the search space has been applied to estimate the parameters of solar cells in an efficient and precise way.In each generation,the INFO utilizes three operations to update the vectors’locations:updating rules,vector merging,and local search.The INFO is applied to estimate the parameters of static models such as single and double diodes,as well as dynamic models such as integral and fractional models.The outcomes of all applications are examined and compared to several recent algorithms.As well as the results are evaluated through statistical analysis.The results analyzed supported the proposed algorithm’s efficiency,accuracy,and durability when compared to recent optimization algorithms.

A Privacy Preserving Federated Learning System for IoT Devices Using Blockchain and Optimization

In this study, a blockchain based federated learning system using an enhanced weighted mean vector optimization algorithm, known as EINFO, is proposed. The proposed EINFO addresses the limitations of federated averaging during global update and model training, where data is unevenly distributed among devices and there are variations in the number of data samples. Using a well-defined structure and updating the vector positions by local searching, vector combining, and updating rules, the EINFO algorithm maximizes the shared model parameters. In order to increase the exploration and exploitation capabilities, the model convergence rate is improved and new vectors are generated through the use of a weighted mean vector based on the inverse square law. To choose validators, miners, and to propagate new blocks, a delegated proof of stake based on the reliability of blockchain nodes is suggested. Federated learning is included into the blockchain to protect nodes from both external and internal threats. To determine how well the suggested system performs in relation to current models in the literature, extensive simulations are run. The simulation results show that the proposed system outperforms existing schemes in terms of accuracy, sensitivity and specificity.

A NEW SVM BASED EMOTIONAL CLASSIFICATION OF IMAGE

How high-level emotional representation of art paintings can be inferred from percep tual level features suited for the particular classes (dynamic vs. static classification)is presented. The key points are feature selection and classification. According to the strong relationship between notable lines of image and human sensations, a novel feature vector WLDLV (Weighted Line Direction-Length Vector) is proposed, which includes both orientation and length information of lines in an image. Classification is performed by SVM (Support Vector Machine) and images can be classified into dynamic and static. Experimental results demonstrate the effectiveness and superiority of the algorithm.

Study of Multistage Project Risk Identification-Assessment Process Based on Objective Orientation

Risk management of projects is about the real time ev aluation and making of decisions proactively in order to maximize the probabilit y of achieving or surpassing the targets set for project objectives. Project objective generally includes three elements: time, cost, quality. Risk occurrin g in the projects will affect these three factors to some various degrees in the end. There are different emphases in each stage and integrated balanced goals b etween the three factors. A large complex engineering project generally consists of several stages each of which has variable objective combinations leading to variable important risks. In order to achieve strategic goals on the schedule under the restriction of lim ited resources, the paper gives the analysis of the so-called risk identificati on-assessment process on the basis of objective orientation. In this paper the set of involved mostly hazards is presented in terms of given objective weight v ector, and so is the model of risk ranking .By reducing the range of risk factor s step by step, risk manager could pay more attention to important ventures and effectively control of them. According to different objective combination at different stages, primary risk f actor sets at different stages are given. With the probability and their various effects to project objectives, evaluation of these sets is made aiming to r educing of the scope of risks and providing decision maker with a better decisio ns support. Successful projects are those, which focus on the relevant business objectives t hroughout the whole process and seek to information integration across project l ife cycle. This paper also introduces the idea of real time process of risk iden tification-assessment and presents a flow chart as a demonstration.

A Pseudo-measurement Modelling Strategy for Active Distribution Networks Considering Uncertainty of DGs

Active distribution networks utilize ad-vanced sensors,communication,and control technologies to achieve flexible and intelligent power distribution management.Reliable state estimation(SE)is crucial for distribution management systems to monitor these net-works.Historically,the scarcity of measurement re-sources has hindered the application of SE technology in distribution networks.Establishing a dependable pseu-do-measurement model for active distribution networks can significantly enhance the feasibility of SE.This paper proposes a pseudo-measurement model that aligns with the actual operating status of the distribution network,considering the uncertainty in output from distributed generations(DGs)such as wind turbines and photovolta-ics.Firstly,it analyzes and models the uncertainty of high-penetration DG output,establishing a reliable out-put model that incorporates the physical characteristics of wind and photovoltaic output.Secondly,it proposes a pseudo-measurement modeling method based on support vector machine(SVM),where the kernel function of the SVM is weighted according to the information entropy of fluctuations in historical operating data.This weighting ensures that the established pseudo-measurement model better reflects the actual operating status of the active distribution network.Finally,a mathematical model for optimizing pseudo-measurement selection is developed,with the minimum state estimation error as the objective function and the observability of the active distribution network system as the constraint.Case studies demon-strate the accuracy and effectiveness of this approach.Index Terms—Distribution network,pseudo-measure-ment,uncertainty of DGs,state estimation,entropy weighting method-support vector machine(EWM-SVM).

STABILITY OF CERTAIN RINGS OF HIGHEST WEIGHT VECTORS

The authors prove the stability of the rings of highest weight vectors of the action of Om x GLn on the complex polynomial rings on Cm,n. As an application, the structure of the rings for m = 3 is determined.

Neurocomputing van der Pauw function for the measurement of a semiconductor's resistivity without use of the learning rate of weight vector regulation

Van der Pauw＇s function is often used in the measurement of a semiconductor＇s resistivity. However, it is difficult to obtain its value from voltage measurements because it has an implicit form. If it can be expressed as a polynomial, a semiconductor＇s resistivity can be obtained from such measurements. Normally, five orders of the abscissa can provide sufficient precision during the expression of any non-linear function. Therefore, the key is to determine the coefficients of the polynomial. By taking five coefficients as weights to construct a neuronetwork, neurocomputing has been used to solve this problem. Finally, the polynomial expression for van der Pauw＇s function is obtained.

Density Weighted Averaging Operator and Application

Exploring structural characteristics implied in initialdecision making information is an important issue in the process of aggregation. In this paper we provide a new family of aggregation operator called density weighted averaging operator（abbreviated as DWA operator）, which carries out the aggregation by classification. In this case, not only the hidden structural characteristics can be identified, some commonly known aggregation operators can also be incorporated into the function of the DWA operator. We further discuss the basic properties of this new operator, such as commutativity, idempotency, boundedness and monotonicity withcertain condition. Afterwards, two important issues related to the DWA operator are investigated, including the arguments partition and the determination of density weights. At last a numerical example regarding performance evaluation of employees is developed to illustrate the using of this new operator.

Defective point data reconstruction based on improved process of morphological operations

In industry,the defective point data often make most surface reconstruction methods suffer from inherent problems that some specific aided information is difficult to obtain. To solve the problem,a novel implicit reconstruction method without any such information is proposed. This approach extends morphological operations into 3D space and provides an improved procedure to construct off-set gradient functions for indirect approximation. By this method,the dual relative functions guarantee a minimal crust surrounding the point data. They can generate a smooth and watertight resulting surface,filling holes and merging overlapping samples reasonably. Compared with other existing methods,the proposed method is better suited to handle defective point clouds in a convenient and efficient manner. The feasibility and effectiveness of the method are demonstrated through a series of practical examples.

Quantum Adjoint Action for Quantum Algebra Uq（f（K,H））

The aim of this paper is to study the adjoint action for the quantum algebra Uq（f（K, H））, which is a natural generalization of quantum algebra Uq（sl2） and is regarded as a class of generalized Weyl algebra..The structure theorem of its locally finite subalgebra F（Uq（f（K, H））） is given.

Personalised medicine with multiple treatments: a PhD thesis abstract

When there is substantial heterogeneity of treatment effectiveness for comparative treatmentselection, it is crucial to identify individualised treatment rules for patients who have heterogeneous responses to treatment. Existing approaches include directly modelling clinical outcomeby defining the optimal treatment rule according to the interactions between treatment andcovariates and outcome weighted approach that uses clinical outcome as weights to maximise atarget function whose value directly reflects correct treatment assignment. All existing articles ofestimating individualised treatment rules are all assuming just two treatment assignments. Herewe propose an outcome weighted learning approach that uses a vector hinge loss to extend estimating individualised treatment rules in multi-category treatments case. The consistency of theresulting estimator is shown. We also demonstrate the performance of our approach in simulationstudies and a real data analysis.

STATIONARY OSCILLATIONS OF LARGE-SCALE NONLINEAR TIME-DELAY SYSTEMS

In this paper,we discuss the existence of stationary oscillations of certain large-scale nonlinear time-delay systems by weighted vector Liapunov functions and obtain a simple sufficiency criterion,which is independent of delays.

Primal Dual Algorithms for the Lexicographically Optimal Base of a Submodular Polyhedron and Its Relation to a Poset Greedoid

We show that for a submodular polyhedron and its dual supermodular polyhedron the exists a unique lexicographically optimal base with respect to a weight vector and they coincide.We also present a dual algorithm to get the lexicograpllically optima base of a submodular polyhedron which works on its dula superlnodular polyhedron.This dual algorithm completely agrees to the algorithm of Morton,G.and von Tandow,R.and Ringwald,K.[1985],where their underlying distributive lattice is a chaill poset greedoid.Finally we show that finding the lesicographically optimal base of a submodular system is essentially equivalent to finding the lexicographically optimal base of a simple submodular system,where its underlying distributive lattice is a poset greedoid.This fact.indicates the importance of greedoids in a further development of submodular system theory.