CONTENTS -- Preface -- Author Index -- Data Mining -- Algorithms for Mining Frequent Sequences Ben Kao and Ming-Hua Zhang -- 1. Introduction -- 2. Problem Definition and Model -- 3. Algorithms -- 3.1. GSP -- 3.2. MFS -- 3.3. Prefixspan -- 3.4. SPADE -- 4. Conclusions -- References -- High Dimensional Feature Selection for Discriminant Microarray Data Analysis Ju-Fu Feng, Jiang-Xin Shi and Qing-Yun Shi -- 1. Introduction -- 2. Introduction of Fisher Linear Discriminant -- 3. Gene Selection Based on Fisher Optimization Model -- 4. Experiment Result and Discussion -- 4.1. MIT AMWALL Data -- 4.2. Colon Cancer Data -- 4.3. Discussion -- 5. Conclusion -- Acknowledgement: -- References -- Clustering and Cluster Validation in Data Mining Joshua Zhe-Xue Huang, Hong-Qiang Rong, Jessica Ting, Yun-Ming Ye and Qi-Ming Huang -- 1. Introduction -- 2. The k-means algorithm family -- 2.1. The k-means algorithm -- 2.2. The k-modes and k-prototypes algorithms -- 2.3. The fuzzy versions of k-means type algorithms -- 3. Visual Cluster Validation -- 3.1. Fastmap algorithm -- 3.2. Cluster Validation with Fastmap -- 4. Classification -- where d is a distance function. -- 5. Conclusions -- References -- Cluster Analysis Using Unidimensional Scaling Pui-Lam hung , Chi-Yin Li and Kin-Nam Lau -- 1. Introduction -- 2. Unidimensional Scaling -- 3. Cluster analysis using UDS -- 4. Conclusion -- References -- Automatic Stock Trend Prediction by Real Time News Gabriel Pui-Cheong Fung, Jeffrey Xu Yu and Wai Lam -- 1. INTRODUCTION -- 2. A NEWS SENSITIVE STOCK TREND PREDICTION SYSTEM -- 2.1. Stock Trend Discovery -- 2.2. Stock Trend Labeling -- 2.3. Article and Trend Alignment -- 2.4. Diflerentiated Feature Weighting -- 2.5. Learning and Prediction -- 3. EVALUATION -- 3.1. Trends Discovery and Labeling -- 3.2. Overall System Performance -- 4. Conclusions -- References.

From Associated Implication Networks to Intermarket Analysis Phil Chi- Wang Tse and Ji-Ming Liu -- 1. Introduction -- 1.1. Related Work -- 1.2. Contributions -- 1.3. Organization of the Paper -- 2. Associated Network Structure and its Discovery -- 2.1. Problem Statement -- 2.2. Discovery of an Associated Network Structure -- 3. The Inference Scheme -- 3.1. Algorithm for Inference with Implication Networks -- 3.2. Algorithm for Inference with Association Rules -- 4. Empirical Validation on Financial Market Analysis -- U.S. Market -- Japanese Market -- 4.1. The Validation Procedure -- 4.2. Settings for the Validation Experiments -- 4.3. Experimental Results -- 4.3.1. The Discovered Network Structure -- 4.3.2. Association Rules Connecting the Networks -- 4.3.3. Experiment to Evaluate the Overall Analysis Performance -- 4.3.4.Experiment to Investigate the Effect of the Number of Observed Nodes on Analysis Performance -- 4.4. Discussion -- 5. Conclusion -- References -- Automating Technical Analysis Philip Leung-Ho Yu, Kin Lam and Sze-Hong Ng -- 1. Introduction -- 2. Identifying Peaks and Troughs in a Price Chart -- 3. Data and Methodology -- 4. Empirical Results -- 5. Conclusion -- References -- Data Modeling -- A Divide-and-Conquer Fast Implementation of Radial Basis Function Networks with Application to Time Series Forecasting Rong-Bo Huang, Yiu-Ming Cheung and Lap-Tak Law -- 1. Introduction -- 2. DCRBF Network -- 2.1. Architecture -- 2.2. Learning Algorithm -- 3. Experimental Results -- 3.1. Experiment 1 -- 3.2. Experiment 2 -- 3.3. Experiment 3 -- 4. Concluding Remarks -- References -- Learning Sunspot Series Dynamics by Recurrent Neural Networks Leong-Kwan Li -- 1. Introduction -- 2. Mathematical Models of Recurrent Neural Networks -- 3. Methodology -- 4. Learning Algorithm of Discrete-time Recurrent Networks -- 5. Results and Discussions.

6. Concluding Remarks -- Acknowledgments: -- References -- Independent Component Analysis: The One-Bit-Matching Conjecture and a Simplified LPM-ICA Algorithm Zhi-Yong Liu, Kai-Chun Chiu and Lei Xu -- 1. Introduction -- 2. A Theorem on the One-Bit-Matching Conjecture -- 3. A Simplified LPM-ICA Algorithm with Only One Free Parameter -- 3.1. Brief Review on LPM-ICA Algorithm -- 3.2. A Simplified LPM-ICA Algorithm -- 4. Experimental Illustration -- 4.1. On Synthetical Data -- 4.2. On Real Data -- 5. Conclusion -- References -- An Higher-Order Markov Chain Model for Prediction of Categorical Data Sequences Wai-Ki Ching, Eric Siu-Leung Fung and Michael Kwok-Po Ng -- 1. Introduction -- 2. Higher-order Markov Chain Models -- 3. Parameters Estimation -- 3.1. Linear Programming Formulation for Estimation of i -- 3.2. An Example -- 4. Some Practical Examples -- 4.1. The DNA Sequence -- 4.2. The Sales Demand Data -- 5. Concluding Remarks -- References -- An Application of the Mixture Autoregressive Model: A Case Study of Modelling Yearly Sunspot Data Kevin Kin-Foon Wong and Chun-Shan Wong -- 1. Introduction -- 2. The Models -- 3. Sunspot Numbers -- 4. Discussion -- References -- Bond Risk and Return in the SSE Long-Zhen Fan -- 1. Introduction -- 2. Yield Curve and Expectations Hypothesis Testing -- 2.1. Notation -- 2.2. Yield Curve in the SSE -- 2.3. Expectations Hypothesis and Testing -- 3. The Forecasting Factors of Bond Excess Returns -- 3.1. Forecasting Bond Excess Returns with Term Structure -- 3.2. Forecasting Short Rate with Term Structure -- 3.3. Principal Component Analysis -- 3.4. A Single Factor for Forecasting Bond Expected Returns -- 3.4 Two factors for Forecasting Bond Expected Returns -- 4. RiskPremia -- 4.1. Calculating Market Price of Risk -- 4.2. Yield Curve Shocks -- 4.3. Empirical Results -- 5. Concluding Remarks -- References.

Mining Loyal Customers: A Practical Use of the Repeat Buying Theory Hing-Po Lo, Xiao-Ling Lu and Zoe Sau-Chun Ng -- 1. Introduction -- 2. Two Models for Repeat Buying -- 3. Two Models for Repeat Buying -- 4. Targeting Loyal Customers for Direct Marketing -- 5. Conclusion -- References.