Microsoft SQL Server

Mining Model Viewer

The Mining Model Viewer allows you to explore and interact with your data mining models within SQL Server Analysis Services. It provides various visualizations and tools to understand the patterns, relationships, and predictions discovered by your models.

Key Features:

• Interactive visualizations tailored to different mining algorithm types.
• Drill-down capabilities to explore data at a granular level.
• Parameter tuning for model exploration.
• Comparison of different models and algorithms.
• Exporting view data and analysis results.

Getting Started:

To use the Mining Model Viewer, you typically connect to an Analysis Services instance using SQL Server Management Studio (SSMS) or a client application that integrates with Analysis Services. Once connected, you can browse your existing mining models and select a viewer appropriate for the model type.

Viewer Controls

Select your mining model and configure viewer options below:

Note: This is a simulated interface. Actual viewer content depends on the selected model and viewer type.

Interactive Visualization Area

Supported Mining Algorithms

The Mining Model Viewer supports a variety of algorithms, each with specific visualization options:

• Decision Trees: Visualize decision paths and prediction rules.
• Clustering: Explore customer segments and their characteristics.
• Association Rules: Discover relationships between items (e.g., market basket analysis).
• Sequence Clustering: Analyze ordered events and patterns.
• Linear Regression: View regression formulas and coefficients.
• Logistic Regression: Examine probability factors for binary outcomes.
• Neural Networks: Visualize layers, weights, and activation functions.
• Time Series: Analyze trends, seasonality, and forecasts.

Data Mining Concepts

Understanding data mining concepts is crucial for effective model building and interpretation:

• Features: Input variables used for prediction.
• Target: The variable you are trying to predict.
• Attributes: Columns within your data sources.
• Cases: Individual records or transactions.
• Patterns: Discoverable structures or relationships in data.
• Predictions: Outputs generated by the mining model.

Advanced Topics

Explore more advanced functionalities and considerations:

• Model properties and parameters.
• Feature relevance analysis.
• Score statistics and accuracy metrics.
• Creating new mining models.
• Integrating with reporting tools like Power BI.