DATA MINING Desktop Survival Guide by Graham Williams
Desktop Survival Project Home List of Figures List of Tables Data Mining Data Mining Data Mining with Rattle Introduction Data Transform Explore A Model Building Framework Unsupervised Modelling Two Class Models Multi Class Models Regression Models Text Mining Evaluation and Deployment Moving into R Troubleshooting R for the Data Miner R Data Graphics in R Understanding Data Preparing Data Building Models Evaluating Models Algorithms Apriori Bagging Bayes Classifier Boosting Cluster Analysis Conditional Trees Hierarchical Clustering K-Means K-Nearest Neighbours Linear Models Logistic Regression Neural Networks Support Vector Machines Text Mining Open Products AlphaMiner Borgelt Data Mining Suite KNime R Rattle Weka Closed Products C4.5 Clementine Equbits Foresight GhostMiner InductionEngine ODM Enterprise Miner Statistica Data Miner TreeNet Virtual Predict Appendicies Glossary Bibliography Index

Usage

GhostMiner supports the extraction of knowledge from data through the following steps:

• Data selection
• Data preprocessing
• Variable selection
• Model learning
• Model analysis
• Selection of the final model

The final output of each GhostMiner project is a model of the knowledge inherent in your data. This model can then be used for decision support.

Figure: Fujitsu GhostMiner interface.

[width=]ghostminer-screen3

GhostMiner includes an intuitive interface for managing projects of arbitrary complexity through a project window and a project tree. Projects store all results of experiments performed so far, automatically adding models created during cross-validation.

Statistical information and charting options are provided to view the data, allowing for the quick detection of outliers. Data may be viewed in its original form or in a standardized/normalized form.

The data can be filtered and ordered in various ways, including......

Numeric and graphical views display variable statistics such as the average, standard deviation, minimum/maximum values and the number of missing values. A facility is provided to show for each class and each variable the distribution of variable values in 2 and 3-dimensional charts, allowing for the identification of potentially useful variables. Two-dimensional projections allow the viewing of data points for particular combination of variables.

A collection of adaptive analytics, including neural networks, neurofuzzy systems, decision trees, and the k-nearest neighbour algorithm are provided. The models can be refined through an ensemble where committees can be composed of arbitrary models or subcommittees of models. GhostMiner introduces K-classifiers which are composed of single models or committees, one for each class.

Various methods for testing the accuracy and efficiency of trained models (cross-validation and -xtest) along with the confusion matrix summarising the model quality are provided.

All these variables are accessible in GhostMiner Developer. After the best model has been selected GhostMiner Analyzer may be used to read it and evaluate new cases. Analyzer displays information about the model and about the data the model has been trained for, without needing to access the tools available in GhostMiner Developer.

Support further development through the purchase of the PDF version of the book.
Brought to you by Togaware.