| Description |
xviii, 308 pages : illustrations ; 24 cm |
| Contents |
Contents note continued: 10.4.Conclusion and Discussion -- References -- pt. III Reliability Analysis -- 11.Reliability in A Feature-Selection Process for Intrusion Detection / Slobodan Petrovic -- 11.1.Introduction -- 11.2.Definition of Reliability in Feature-Selection Process -- 11.3.Generic Feature Selection Measure -- 11.3.1.Definitions -- 11.3.2.Polynomial Mixed 0-1 Fractional Programming -- 11.3.3.Optimization of the GeFS Measure -- 11.4.Experiment -- 11.4.1.Data Sets -- 11.4.2.Experimental Settings -- 11.4.3.Experimental Results -- 11.5.Conclusions -- References -- 12.The Impact of Sample Size and Data Quality to Classification Reliability / Honghua Dai -- 12.1.Introduction -- 12.2.The original data sets and the data set with introduced errors -- 12.3.The examination of the impact of Low Quality Data to the reliability of discovered knowledge -- 12.4.Can we trust the knowledge discovered from a small data set? -- |
|
Contents note continued: 12.5.A Comparison of a traditional classifier learner and an inexact classifier learner -- 12.6.Conclusion and future work -- References -- 13.A Comparative Analysis of Instance-based Penalization Techniques for Classification / Evgueni Smirnov -- 13.1.Introduction -- 13.2.Penalization in learning -- 13.3.Three instance-based classification methods -- 13.4.Alternative specifications -- 13.5.Estimation -- 13.5.1.Support Vector Machines -- 13.5.2.Support Hyperplanes -- 13.5.3.Nearest Convex Hull classifier -- 13.5.4.Soft Nearest Neighbor -- 13.6.Comparison results -- 13.7.Conclusion -- References -- 14.Subsequence Frequency Measurement and its Impact on Reliability of Knowledge Discovery in Single Sequences / Honghua Dai -- 14.1.Introduction -- 14.2.Preliminaries -- 14.3.Previous Frequency Metrics and Their Properties -- 14.3.1.Definitions of Seven Frequency Metrics -- 14.3.2.Properties -- |
|
Contents note continued: 14.4.Inherent Inaccuracies and Their Impacts on Discovered Knowledge -- 14.4.1.Frequent Episodes -- 14.4.2.Episode Rules -- 14.4.3.Findings -- 14.5.Suggestions and A New Frequency Metric -- 14.5.1.Restriction of Window Width -- 14.5.2.Strict Anti-Monotonicity -- 14.5.3.A New Frequency Metric and Its Computation -- 14.6.Empirical Evaluation -- 14.7.Conclusion -- References -- pt. IV Reliability Improvement Methods -- 15.Improving Reliability of Unbalanced Text Mining by Reducing Performance Bias / Honghua Dai -- 15.1.Introduction -- 15.2.Reducing Bias On Majority Class -- 15.2.1.Preliminaries -- 15.2.2.Feature Selection Fish-Net -- 15.3.Reducing Bias On Minority Class -- 15.3.1.Learning Stage -- 15.3.2.Evaluation Stage -- 15.3.3.Optimization Stage -- 15.4.Experimental Results -- 15.4.1.Data Set -- 15.4.2.Results on Inexact Field Learning -- 15.4.3.Results on One-class Classifiers -- 15.5.Conclusion -- References -- |
|
Contents note continued: 16.Formal Representation and Verification of Ontology Using State Controlled Coloured Petri Nets / Xi-Zhao Wang -- 16.1.Introduction -- 16.2.Modeling Ontology by SCCPN -- 16.2.1.Formal Formulation of Ontology -- 16.2.2.SCCPN Notations and Interpretations -- 16.2.3.Formal Definition of SCCPN -- 16.3.Ontology Inference in SCCPN -- 16.3.1.Markings for Representation of Inference -- 16.3.2.Inference Mechanisms for Different Relation Types -- 16.3.3.An Illustrative Example -- 16.4.Potential Anomalies in Ontology and Formal Verification -- 16.4.1.Redundancy -- 16.4.2.Circularity -- 16.4.3.Contradiction -- 16.5.Performance Analysis -- 16.5.1.Modeling Ontology by SCCPN -- 16.5.2.Complexity Analysis of Ontology Verification -- 16.6.Conclusion -- References -- 17.A Reliable System Platform for Group Decision Support under Uncertain Environments / James N.K. Liu -- 17.1.Introduction -- 17.2.Group Multiple Criteria Decision Analysis -- |
|
Contents note continued: 17.2.1.Multiple Criteria Decision Making -- 17.2.2.General Problem Model of Group MCDM -- 17.3.Uncertainty Multiple Criteria Decision Analysis -- 17.3.1.Stochastic MCDM -- 17.3.2.Fuzzy MCDM -- 17.3.3.Rough MCDM -- 17.4.UGDSS Framework -- 17.4.1.Uncertainty Group Decision Process and System Structure -- 17.4.2.UGDSS Architecture -- 17.4.3.Knowledge-related System Designs -- 17.5.Conclusion -- References |
|
Contents note continued: 2.2.3.Reliable machine learning from data streams -- 2.3.Estimating reliability of individual streaming predictions -- 2.3.1.Preliminaries -- 2.3.2.Reliability estimates for individual streaming predictions -- 2.3.3.Evaluation of reliability estimates -- 2.3.4.Abalone data set -- 2.3.5.Electricity load demand data stream -- 2.4.Correcting individual streaming predictions -- 2.4.1.Correcting predictions using the CNK reliability estimate -- 2.4.2.Correcting predictions using the Kalman filter -- 2.4.3.Experimental evaluation -- 2.4.4.Performance of the corrective approaches -- 2.4.5.Statistical comparison of the predictions' accuracy -- 2.5.Conclusions -- References -- 3.Error Bars for Polynomial Neural Networks / Evgueni Smirnov -- 3.1.Introduction -- 3.2.Genetic Programming of PNN -- 3.2.1.Polynomial Regression -- 3.2.2.Tree-structured PNN -- 3.2.3.Weight Learning -- 3.2.4.Mechanisms of the GP System -- 3.3.Sources of PNN Deviations -- |
|
Contents note continued: 3.4.Estimating Confidence Intervals -- 3.4.1.Delta Method for Confidence Intervals -- 3.4.2.Residual Bootstrap for Confidence Intervals -- 3.5.Estimating Prediction Intervals -- 3.5.1.Delta Method for Prediction Intervals -- 3.5.2.Training Method for Prediction Bars -- 3.6.Conclusion -- References -- pt. II Reliable Knowledge Discovery Methods -- 4.Robust-Diagnostic Regression: A Prelude for Inducing Reliable Knowledge from Regression / Honghua Dai -- 4.1.Introduction -- 4.2.Background of Reliable Knowledge Discovery -- 4.3.Linear Regression, OLS and Outliers -- 4.4.Robustness and Robust Regression -- 4.4.1.Least Median of Squares Regression -- 4.4.2.Least Trimmed Squares Regression -- 4.4.3.Reweighted Least Squares Regression -- 4.4.4.Robust M (GM)- Estimator -- 4.4.5.Example -- 4.5.Regression Diagnostics -- 4.5.1.Examples -- 4.6.Concluding Remarks and Future Research Issues -- References -- 5.Reliable Graph Discovery / Honghua Dai -- 5.1.Introduction -- |
|
Contents note continued: 5.2.Reliability of Graph Discovery -- 5.3.Factors That Affect Reliability of Graph Discovery -- 5.4.The Impact of Sample Size and Link Strength -- 5.5.Testing Strategy -- 5.6.Experimental Results and Analysis -- 5.6.1.Sample Size and Model Complexity -- 5.7.Conclusions -- References -- 6.Combining Version Spaces and Support Vector Machines for Reliable Classification / Ida Sprinkhuizen-Kuyper -- 6.1.Introduction -- 6.2.Task of Reliable Classification -- 6.3.Version Spaces -- 6.3.1.Definition and Classification Rule -- 6.3.2.Analysis of Version-Space Classification -- 6.3.3.Volume-Extension Approach -- 6.4.Support Vector Machines -- 6.5.Version Space Support Vector Machines -- 6.5.1.Hypothesis Space -- 6.5.2.Definition of Version Space Support Vector Machines -- 6.5.3.Classification Algorithm -- 6.6.The Volume-Extension Approach for VSSVMs -- 6.7.Experiments -- 6.8.Comparison with Relevant Work -- 6.8.1.Bayesian Framework -- 6.8.2.Typicalness Framework -- |
|
Contents note continued: 6.9.Conclusion -- References -- 7.Reliable Ticket Routing in Expert Networks / Nikos Anerousis -- 7.1.Introduction -- 7.2.Related Work -- 7.3.Preliminaries -- 7.4.Generative Models -- 7.4.1.Resolution Model (RM) -- 7.4.2.Transfer Model (TM) -- 7.4.3.Optimized Network Model (ONM) -- 7.5.Ticket Routing -- 7.5.1.Ranked Resolver -- 7.5.2.Greedy Transfer -- 7.5.3.Holistic Routing -- 7.6.Experimental Results -- 7.6.1.Data Sets -- 7.6.2.Model Effectiveness -- 7.6.3.Routing Effectiveness -- 7.6.4.Robustness -- 7.7.Conclusions and Future Work -- References -- 8.Reliable Aggregation on Network Traffic for Web Based Knowledge Discovery / Wanchun Dou -- 8.1.Introduction -- 8.2.The Reliability of Network Traffic Information -- 8.3.Aggregation Functions -- 8.4.Information Theoretical Notions of Distance -- 8.5.Performance Comparison for Information Distances -- 8.6.Summary -- References -- |
|
Contents note continued: 9.Sensitivity and Generalization of SVM with Weighted and Reduced Features / Li-wei Jia -- 9.1.Introduction -- 9.2.Background -- 9.2.1.The Classical SVM Regression Problem -- 9.2.2.Rough Set SVM Regression -- 9.2.3.Grey Correlation Based Feature Weighted SVM Regression -- 9.3.Experimental Results and Analysis -- 9.3.1.Data Collection -- 9.3.2.Data Pre-processing -- 9.3.3.Kernel Function Selection and Parameter Selection -- 9.3.4.The Experiments -- 9.4.Conclusions and Future Works -- References -- 10.Reliable Gesture Recognition with Transductive Confidence Machines / Youri van Pinxteren -- 10.1.Introduction -- 10.2.Methods -- 10.2.1.Transductive Confidence Machines -- 10.2.2.The TCM-kNN algorithm and its complexity -- 10.2.3.The NicIcon dataset and DTW-based trajectory matching -- 10.2.4.Modification -- 10.3.Experiments and Results -- 10.3.1.Modified TCM algorithm -- 10.3.2.Writer Dependent Set -- 10.3.3.Writer Independent Set -- 10.3.4.Error Samples -- |
|
Machine generated contents note: pt. I Reliability Estimation -- 1.Transductive Reliability Estimation for Individual Classifications in Machine Learning and Data Mining / Matjaz Kukar -- 1.1.Introduction -- 1.2.Related work -- 1.2.1.Transduction -- 1.3.Methods and materials -- 1.3.1.Typicalness -- 1.3.2.Transductive reliability estimation -- 1.3.3.Merging the typicalness and transduction frameworks -- 1.3.4.Meta learning and kernel density estimation -- 1.3.5.Improving kernel density estimation by transduction principle -- 1.3.6.Testing methodology -- 1.4.Results -- 1.4.1.Experiments on benchmark problems -- 1.4.2.Real-life application and practical considerations -- 1.5.Discussion -- References -- 2.Estimating Reliability for Assessing and Correcting Individual Streaming Predictions / Igor Kononenko -- 2.1.Introduction -- 2.2.Background -- 2.2.1.Computation and utilization of prediction reliability estimates -- 2.2.2.Correcting individual regression predictions -- |
| Notes |
Formerly CIP. Uk |
| Bibliography |
Includes bibliographical references and index |
| Subject |
Data mining.
|
| Author |
Dai, Honghua.
|
|
Smirnov, Evgueni
|
| ISBN |
1461419026 (hbk.) |
|
9781461419020 (hbk.) |
|
