Predictive Analytics Data Mining Tools

Predictive modeling and Data Mining techniques have many applications within business intelligence processes. For example, credit risk, fraud detection, quality control, warranty analysis, customer segmentation, customer lifetime value and cross-sell/up-sell analyses all require advanced analytics.

InforSense provides an extensible toolset for predictive modeling methods including Decision Tree Induction, Decision Rule Induction, Naïve Bayes Classification, Logistic Regression, Neural Net, Support Vector Machines and an Ensemble Classification approach. It also provides extensions for multivariate analysis techniques including Principal Component Analysis (PCA), Partial Least Squares (PLS), Linear and Polynomial Regression and Neural Net Regression. Extensions for clustering analysis include K-Means Clustering, Hierarchical Clustering and Expectation Maximization clustering.

Each of these data mining tools are supported by advanced data preparation methods to gain optimal results from specific algorithms and multiple model diagnostic methodologies including cross-validation and boot-strapping approaches for model evaluation.

In addition, the system offers interfaces to Oracle-based analytics components, including Oracle Data Mining and Oracle statistics and to various statistical analysis tools including R and Matlab.

InforSense also provides interactive tools to guide experts through the development of predictive models including:

• Classification Studio for building and optimization of classification models for predictive data mining tasks.
• Clustering Studio for building and optimization of clustering models for segmentation analysis and descriptive data mining tasks.
• Interactive Decision Tree to enable experts to guide the construction of decision tree models.