Jiali Feng

Research on multi-agent comprehensive evaluation model based on attribute coordinate

Some researchers have made great achievements in both theoretical study and practical applications in comprehensive evaluation models based on an attribute coordinate, but they failed to make breakthroughs in multi-agent evaluation. This paper firstly constructs barycentric coordinate points based on a multi-agent and then attains the psychological curve of multi-agent evaluation using the Lagrange interpolation formula. It further builds up the multi-agent evaluation model of attribute coordinates to achieve superior functionality of multi-agent evaluation.

Research and Implementation of Image Encryption Algorithm Based on Zigzag Transformation and Inner Product Polarization Vector

Zigzag transformation is a kind of scrambling algorithm with low time complexity and good scrambling effect. And inner product polarization vector algorithm is a new encryption algorithm deduced from Attribute Theory. Here both above are combined by first making use of improved Zigzag transformation to have the image scrambled, next encrypting the scram-bled image by way of inner product polarization vector. Thus the purpose to change both positions and values of the image pixels can be attained. Simulations apply the algorithm to RGB colorful images. And the analysis of the security of algorithm is also given.

Attribute Grid Computer based on Qualitative Mapping and its application in pattern Recognition

A new kind of Computer, called Attribute Grid Computer based on Qualitative Mapping is presented in this paper, It is shown that a series of intelligent methods, such as Production System, Artificial Neural Network, and Support Vector Machine can be fused in the framework of qualitative criterion transformation of qualitative mapping and can be implemented by attribute grid computer. And some examples of application in pattern recognition are given too.

A kind of synthetic evaluation method based on the attribute computing network

Based on input and output relationship of Qualitative Mapping(QM), the attribute computing network model has been created. It brings forward a kind of computing method using input to adjust qualitative benchmark of attribute network, which makes it possible to achieve pattern recognition. Now the new attribute computing network model combined pattern recognition with synthetic evaluation is established. Firstly qualitative benchmarks of indexes are gotten by boundary study, and then by way of marking, preference for indexes is obtained, and lastly a set of satisfactory degrees for indexes is computed and outputted in descending sequence which ameliorates the effect of old satisfactory degree. Finally the simulation experiment is carried out to validate the theoretical model.

Qualitative Mapping, Criterion Trasformation and Artificial Neuron

A mathematical model for describing the disciplinarian that the true value of property p(x) varies according to its qualitative criterion [alpha, beta], called the qualitative mapping taup (x,[alpha, beta]) is presented in this paper, and the inner product transformation of qualitative criterion [alpha, beta], denoted by w_[alpha, beta] is discussed. It is shown that an artificial neuron is a special qualitative mapping

Attribute Computing Network Induced by Linear Transformation and Granular Transformation of Qualitative Criterion

It is shown that not only the qualitative criterion [alpha,beta] of qualitative mapping is a bridge between expert system and artificial neural network, the qualitative mapping and the artificial neuron can be defined each other, but support vector machine can be also induced by the granular transformation of qualitative criterion, the qualitative mapping is a mathematical model by which some of artificial intelligent methods can be fused and unified together, and a kind of artificial fused model: attribute computing network induced by qualitative mapping is presented.

Study on Updating Algorithm of Attribute Coordinate Evaluation Model

Evaluation model based on attribute coordinate has made some achievements in both theoretical research and practical applications. However, if the new evaluation samples are added, the evaluation model needs to be reconstructed rather than the dynamic updating. Almost no progress has been made on how to dynamically update the evaluation model. Thus, this paper puts forward a dynamic updating algorithm based on barycentric coordinates and satisfaction function to effectively solve this problem. The experiment results show the reasonability and effectiveness of this algorithm.

A Ranking Model Based on Credit for Social News Website

Social news websites were popular for a time, but recently the traffics of some social news websites have been slowing down, even some of them went out of business. It’s caused partly by the emergence of massive spam news and the news publication algorithm heavily relying on the ranking of senior users while ignoring the ranking of ordinary users. In order to solve these problems, this paper proposes a ranking model based on credit. On the technology, utilizes the qualitative mapping as the credit threshold activation model, uses the K-means method for classifying the users and attribute ranging model to calculate the value of the user credit. The simulation experiment shows this model can effectively reduce the spam news and pay more attention on regular users.

A Topos Induced by Qualitative Mapping

It is shown that the Reasoning Lattice-Monoid Category L-M(Au, Ù, →) of the attribute set Au of object u can be constructed by the conjunction Ù between two attributes, and a Qualitaive Mapping Topos TQM(Au, *, →) of the category L-M(Au, *, →) can be induced by a Qualitative Mapping t(x, [a, b]) with qualitative criterion transformation T.

Predicting gene function using few positive examples and unlabeled ones

BackgroundA large amount of functional genomic data have provided enough knowledge in predicting gene function computationally, which uses known functional annotations and relationship between unknown genes and known ones to map unknown genes to GO functional terms. The prediction procedure is usually formulated as binary classification problem. Training binary classifier needs both positive examples and negative ones that have almost the same size. However, from various annotation database, we can only obtain few positive genes annotation for most offunctional terms, that is, there are only few positive examples for training classifier, which makes predicting directly gene function infeasible.ResultsWe propose a novel approach SPE_RNE to train classifier for each functional term. Firstly, positive examples set is enlarged by creating synthetic positive examples. Secondly, representative negative examples are selected by training SVM(support vector machine) iteratively to move classification hyperplane to a appropriate place. Lastly, an optimal SVM classifier are trained by using grid search technique. On combined kernel ofYeast protein sequence, microarray expression, protein-protein interaction and GO functional annotation data, we compare SPE_RNE with other three typical methods in three classical performance measures recall R, precise P and their combination F: twoclass considers all unlabeled genes as negative examples, twoclassbal selects randomly same number negative examples from unlabeled gene, PSoL selects a negative examples set that are far from positive examples and far from each other.ConclusionsIn test data and unknown genes data, we compute average and variant of measure F. The experiments showthat our approach has better generalized performance and practical prediction capacity. In addition, our method can also be used for other organisms such as human.