## Proceedings of the Fifth SIAM International Conference on Data MiningThe Fifth SIAM International Conference on Data Mining continues the tradition of providing an open forum for the presentation and discussion of innovative algorithms as well as novel applications of data mining. Advances in information technology and data collection methods have led to the availability of large data sets in commercial enterprises and in a wide variety of scientific and engineering disciplines. The field of data mining draws upon extensive work in areas such as statistics, machine learning, pattern recognition, databases, and high performance computing to discover interesting and previously unknown information in data. This conference results in data mining, including applications, algorithms, software, and systems. |

### What people are saying - Write a review

We haven't found any reviews in the usual places.

### Contents

A Rondom Wolks Perspective on Moximizing Satisfaction and Profit | 12 |

Surveying Doto for Patchy Structure | 20 |

Chris Ding and Jieping | 32 |

Mining Frequent itemsets from Data Streams with a TimeSensitive Sliding Window | 68 |

PrivacyPreserving Classification of Customer Dato without Loss of Accuracy | 92 |

A Fedsible Approach for Inverse | 103 |

On Varioble Constraints in Privacy Preserving Dato Mining | 115 |

Clustering with ModelLevel Constroints | 126 |

Variotional Learning for NoisyOR Component Anolysis | 370 |

A New Stotistic for the Structurdl Breck Defection in Time Series | 392 |

Efficient Mining of Moximal Sequentiol Potterns Using Multiple Somples | 415 |

Correlation Clustering for Leorning Mixtures of Cononical Correlotion Models | 439 |

Mining loeberg Cubes from Doto Worehouses | 461 |

Decision Tree Induction in High Dimensional Hierorchicolly Distributed Dotoboses | 466 |

Sparse Fisher Discriminant Andlysis for Computer Aided Detection | 476 |

Mcking Doto Mining Models Useful to Model Nonpoying Customers of Exchange Corriers | 486 |

Fedsibility issues and the kMeans Algorithm | 138 |

A Cutting Algorithm for the Minimum SumofSquared Error Clustering | 150 |

Dynomic Clossification of Defect Structures in Molecular Dynomics Simulation Doto | 161 |

Simultaneous Mining of Positive and Negotive Spotial | 173 |

Finding Young Stellar Populations in Elliptical Galaxies from Independent Components | 183 |

Hybrid Attribute Reduction for Classification Bosed on a Fuzzy Rough Set Technique | 195 |

Contents | 205 |

Lazy Learning for Classification Based on Query Projections | 227 |

DepthFirst Nonderivable itemset Mining | 250 |

A Spectral Clustering Approach to Finding Communities in Graphs | 274 |

Learning to Refine Ontology for a New Web Site Using a Bayesian Approach | 298 |

Exploiting Geometry for Support Vector Mochine Indexing | 322 |

A Lodd Shedding Scheme for Classifying Doto Streams | 346 |

Cluster Validity Anolysis of Alternotive Results from Multiobjective Optimization | 496 |

A Cose Study on Face Recognition | 511 |

On Clustering Bindry Doto | 526 |

Pushing Fedture Selection Ahedd of Join | 536 |

Discretizotion Using Successive Pseudo Deletion of Moximum Information Goin | 546 |

The Best Nurturers in Computer Science Resedrch | 566 |

Confents | 571 |

NearNeighbor Search in Pattern Distance Spaces | 586 |

On the Equivalence of Nonnegotive Motrix Foctorization and Spectral Clustering | 606 |

Correcting Sampling Bios in Structural Genomics through iterative Selection | 621 |

WFIM Weighted Frequent itemset Mining with a Weight Range and a Minimum weight | 636 |

### Common terms and phrases

accuracy algorithm analysis applied approach approximation association assume attribute bound called candidate classification closed clustering compute concepts condition Conference consider constraints contains correlation corresponding count data mining data streams database dataset decision defined denoted derived detection distance distribution effective efficient error estimation example experiments Figure frequent function given graph increase instances itemsets Knowledge label learning linear matrix means measure method node Note objects observed obtained optimal parameters patterns performance points positive prediction present probability problem proposed pruning query random records References represent respectively rules sample scheme selection sequence shown similar solution space statistics step structure subset Table techniques tion transactions tree vector weights