By Topic

Business Intelligence and Financial Engineering, 2009. BIFE '09. International Conference on

Date 24-26 July 2009

Filter Results

Displaying Results 1 - 25 of 209
  • [Front cover]

    Page(s): C1
    Save to Project icon | Request Permissions | PDF file iconPDF (389 KB)  
    Freely Available from IEEE
  • [Title page i]

    Page(s): i
    Save to Project icon | Request Permissions | PDF file iconPDF (9 KB)  
    Freely Available from IEEE
  • [Title page iii]

    Page(s): iii
    Save to Project icon | Request Permissions | PDF file iconPDF (54 KB)  
    Freely Available from IEEE
  • [Copyright notice]

    Page(s): iv
    Save to Project icon | Request Permissions | PDF file iconPDF (109 KB)  
    Freely Available from IEEE
  • Table of contents

    Page(s): v - xviii
    Save to Project icon | Request Permissions | PDF file iconPDF (185 KB)  
    Freely Available from IEEE
  • Preface

    Page(s): xix - xx
    Save to Project icon | Request Permissions | PDF file iconPDF (65 KB)  
    Freely Available from IEEE
  • Organizing Committee

    Page(s): xxi
    Save to Project icon | Request Permissions | PDF file iconPDF (48 KB)  
    Freely Available from IEEE
  • Program Committee

    Page(s): xxii - xxiv
    Save to Project icon | Request Permissions | PDF file iconPDF (71 KB)  
    Freely Available from IEEE
  • Workshop Co-chairs

    Page(s): xxv
    Save to Project icon | Request Permissions | PDF file iconPDF (50 KB)  
    Freely Available from IEEE
  • list-reviewer

    Page(s): xxvi - xxvii
    Save to Project icon | Request Permissions | PDF file iconPDF (49 KB)  
    Freely Available from IEEE
  • An Algorithm for Determining Neural Network Architecture Using Differential Evolution

    Page(s): 3 - 7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (287 KB) |  | HTML iconHTML  

    Artificial neural networks (ANNs) have been applied to a variety of classification and learning tasks. The use of evolutionary algorithms (EA) as one of the fastest, robust and efficient global search techniques has allowed different properties of artificial neural networks to be evolved. This paper proposes the possibility of using differential evolution for determining an ANN architecture (DNNA). We explain how to use differential evolutionpsilas application for determining an ANN architecture. The approach we describe is innovative and has only been successfully applied and implemented for the first time, although the idea of differential evolution has been applied in various fields since the last decade. In this work, we proposed an algorithm based on differential evolution that uses a minimum number of user specified parameters in determining an ANN architecture. By using backpropagation algorithm to train the ANN architecture partially during the evolution process, DNNA is evaluated on five benchmark classification problems, namely, Cancer, Diabetes, Heart Disease, Thyroid, and the Australian Credit Card problem. Through performance analysis and simulation studies, we show that DNNA can produce ANN architecture with good generalization abilities, but with less number of training cycles when compared with an evolutionary programming approach and standard backpropagation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Predicting China's Energy Consumption Using Artificial Neural Networks and Genetic Algorithms

    Page(s): 8 - 11
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (288 KB) |  | HTML iconHTML  

    In this work, artificial neural networks (ANN) based on genetic algorithm (GA) have been developed to predict energy consumption in China. The numbers of neurons in the hidden layer, the momentum rate and the learning rate are determined using the genetic algorithm. The inputs to the artificial neural networks model are four variables, namely, gross domestic product, industrial structure, total population and technology progress. It is verified that genetic algorithm could find the optimal architecture and parameters of the back-propagation algorithm. In addition, the artificial neural network model based genetic algorithm is tested and the results indicate that the energy consumption in China can be efficiently forecasted by this model. Compared with a network in which the ANN calibration is done using a trial-and-error approach, it can be found that this model can improve prediction accuracy. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Stock Bubbles' Nature: A Cluster Analysis of Chinese Shanghai a Share Based on SOM Neural Network

    Page(s): 12 - 16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (291 KB) |  | HTML iconHTML  

    The stock market bubbles present different properties in different economic environments and stages, and their impacts on the economic system are varied. In this paper, self organizing map (SOM) and principal component analysis (PCA) were employed to determine the property of the stock bubbles in Shanghai stock market from Jan-2000 to Apr-2008. The nature of the bubbles was interpreted by factor analysis from the aspects of macroeconomic, stock marketpsilas speculative intensity and dilatation. The factors analyses of bubbles explained the bubblespsila nature by the characters of macroeconomic, stock market speculative intensity and expansion. The outcome demonstrates that SOM may help to determine the property of the bubbles in stock market. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The Application of Artificial Neural Networks in Risk Assessment on High-Tech Project Investment

    Page(s): 17 - 20
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (226 KB) |  | HTML iconHTML  

    Investment risks assessment of high-tech projects is a more complex process, involving various factors and it is not entirely the linear relationship between influencing factors and measurement results. Artificial neural network (ANN) has a strong nonlinear mapping ability, with strong learning ability and high classification and prediction accuracy. The paper applied ANN to establish a new risk assessment model of high-tech project investment and used MATLAB software to carry out example simulations respectively with BP neural network model and RBF neural network model. The results showed that it is effective to apply ANN to assess the high-tech project investment risk and RBF network is more suitable than BP network. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • IPO Pricing of SME Based on Artificial Neural Network

    Page(s): 21 - 24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (235 KB) |  | HTML iconHTML  

    The pricing method of new shares issuing based on artificial neural network is studied in this paper. A three-layer neural network model is established and simulation tests are carried out. It shows that the BP network model established fits well with the real first day's closing price of stock and greatly improves the IPO pricing. It provides a new way to investors for forecasting IPO price of small and medium enterprise. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Appraisal of High-Tech Zone Technology Innovation Ability Based on BP Neural Network

    Page(s): 25 - 28
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (235 KB) |  | HTML iconHTML  

    High-tech industry zone play a important role in the global economic system with the continuously new knowledge innovation, itpsilas becoming the main motion of regional economic structure optimization and competition. The technology ability and diffuse effect of High-tech zone influence itself and regional economic development greatly. This paper design High-tech zone technology innovation ability appraisal index system, and take integrated appraisal with BP neural network, a improved method is put forward to identification mode by Euclidean distance, and appraise a true sample with it. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The Finger Movement Identification Based on Fuzzy Clustering and BP Neural Network

    Page(s): 29 - 33
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (252 KB) |  | HTML iconHTML  

    The classification and identification technology plays an important role in the research of brain-computer interface (BCI) systems. In this paper, we do fuzzy clustering disposal for the multi-channel electroencephalogram (EEG) during finger movement at first according to event-related desynchronization phenomena (ERD) in the event-related EEG. Then we classify signal-trial EEG with the feature extracted from EEG based on common spatial subspace decomposition (CSSD) algorithm. The averaged classification accuracy achieves 94.58% in the course of testing on the data from four subjects. Experiment results show that the combination of fuzzy clustering and neural network can greatly improve the rate of identification for finger movement. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Application of Hopfield Neural Network in Target Selection of Mergers and Acquisitions

    Page(s): 34 - 37
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (263 KB) |  | HTML iconHTML  

    Target selection is one of the most important steps of during the process of mergers and acquisitions. Hopfield neural network is very strong in pattern recognition which can simulate the criteria of acquirer and remind it. The network model overcomes the shortcomings of classic statistic and fuzzy models and embodies the requirements of acquirer. Demonstration shows that Hopfield network is an effective tool to choose targets with good performance if the standard of mergers and acquisition is feasible and reasonable. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modelling and Prediction of the CNY Exchange Rate Using RBF Neural Network

    Page(s): 38 - 41
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (326 KB) |  | HTML iconHTML  

    The CNY exchange rates can be viewed as financial time series which are charactered by high uncertainty, nonlinearity and time-varying behavior. Predictions for exchange rates of GBP-CNY and USD-CNY were carried respectively by means of RBF neural network forecasters. The detailed designs for architectures of RBF neural network models, transfer functions of the hidden layer nodes, input vectors and output vectors were made with many tests. Experimental results show that the performance of RBF neural networks for forecasting CNY spot exchange rates is acceptable and effective. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Method of Gear Fault Diagnosis Based on CWT and ANN

    Page(s): 42 - 45
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (249 KB) |  | HTML iconHTML  

    Aimed at the engine rotor fault, a new diagnosis method based on Wavelet Transform and artificial neural network (ANN) is proposed. Firstly, according to the wavelet transform theories, the original signals are sampling repeatedly, and the continuous wavelet transform (CWT) is used for the signals sampled. Afterward, the obtained signals are decomposed to fixed layer so as to obtain the frequency band characteristics of the original signals. So the traditional spectrum features are extracted, and the feature vector is obtained. Second, we use ANN technique to diagnose the selected features intelligently. The results adequately prove that the methods of feature extraction and feature selection advanced in this paper are rational and effective. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Research of Dispatching Method in Elevator Group Control System Based on Traffic Mode Identify

    Page(s): 46 - 49
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (259 KB) |  | HTML iconHTML  

    Elevator group control system (EGCS) with multi-objective, stochastic and nonlinear characteristics is a complex optimization system. After analyzing characteristic of typical traffic mode of elevator. This paper proposed a new simulation platform of an elevator group control system implemented in C# using the fuzzy-neural network technology. The result of simulation shows that this method realizes reasonable elevator dispatching under various passenger traffic conditions and indicates the validity of this method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Power Futures Price Forecasting Based on RBF Neural Network

    Page(s): 50 - 52
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (211 KB) |  | HTML iconHTML  

    In order to forecast power futures price exactly, a radial basis function neural network (RBF NN) method is used in this paper. The RBF NN method has the advantages of rapid training, generality and simplicity over feed-forward neural network. The data of Nordic electricity market is adopted for case analysis. Empirical results reveal that the RBF NN method has a more accurate result than back-propagation neural network (BP NN) method. The RBF NN method can effectively forecast power futures price. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Personal Credit Rating Assessment for the National Student Loans Based on Artificial Neural Network

    Page(s): 53 - 56
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (244 KB) |  | HTML iconHTML  

    National student Loans are the use of the financial means to improve the college subsidy policy. State Student Loan is a personal credit loan, but the personal credit assessment system of commercial banks could not make a correct assessment for a college Studentpsilas credit rating because the students have no records about their credit. To avoid the credit risk, it must to establish a rational credit assessment system for college Students. Artificial neural network can simulate, to some extent, how neural network in human brain deals with, searches and stores information. With its self-learning, self-organizing, adaptive and nonlinear dynamic handling characteristics, a Back Propagatio neural network was developed to evaluate the credit rating about a college student. 16 samples was used for network training and testing by MATLAB. The maximum value of the error between the prediction value of the network and actual value is only 2.92%. Simulation results demonstrate that the algorithm developed is fairly efficient for the assessment about the college studentpsilas personal credit situation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Integration of Unascertained Method and Neural Networks in Financial Early Warning

    Page(s): 57 - 60
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (254 KB) |  | HTML iconHTML  

    Unascertained system that imitates the human brain's thinking logical is a kind of mathematical tools used to deal with imprecise and uncertain knowledge. Artificial neural network that imitates the function of human neurons may function as a general estimator, mapping the relationship between input and output. Combination of these two methods were made attemption in financial early warning in this paper. Application case shows that combines unascertained systems with feedforward artificial neural networks can obtain more reasonable and more advantage of nonlinear mapping that can handle more complete type of data. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Anti-dumping Early-Warning Model Based on Entropy Weight and SOM

    Page(s): 61 - 64
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (243 KB) |  | HTML iconHTML  

    A new anti-dumping early-warning system for the export of China's products is presented, which is based on entropy weight method and SOM neutral network. It is different from traditional modeling methods. We can acquire the indexpsila weight by entropy method, and then, the indexes in which the weight are decided are used to develop classification rules and train SOM nerve network. The result of the positive research indicated that this system is very valid for anti-dumping prediction and it will have a good application prospect in this area. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.