|
Predictive Modeling & Data Mining |
|
Predictive modeling from SwiftKnowledge empowers organizations to run “what if” analyses where variables are modified to determine their impact on the business. These exercises help discover, for instance, which variables hold the greatest impact, which variables interact as a group, and which variables can be controlled or mitigated.
SwiftKnowledge’s data mining capabilities allow companies to discover hidden patterns in data – an increasingly important tool for transforming “data” into “information.” It is commonly used in a wide range of profiling practices, such as grouping customers into different purchasing patterns, or clusters, in order to determine specific marketing messaging.
The Power of Predictive Modeling and Data Mining from SwiftKnowledge
- Easily run “what if” and change scenarios using web-based user interface
- Design custom dashboards/mashboards and reports using easy drag-and-drop, filter and drill-through controls
- Schedule reports to be delivered directly to email inboxes and utilize automated alerts to proactively push out critical, actionable information
- Select from multiple output formats – grids, charts, meters, maps and PDFs – to draw immediate attention to key business metrics
Sample Applications
- Finance – An analyst runs real estate loan stress tests featuring comprehensive sensitivity analysis scenarios with alternate variables such interest rates, capitalization rates and/or occupancy rates to determine loan risk and debt service coverage (DSC) for a single loan or groups of loans.
- Marketing – A marketer uses data mining algorithms to determine which products tend to be purchased by customers with particular attributes (age, gender, location), then combines that information with predictive algorithms to recommend with a higher degree of success the most appropriate products for a single individual or a group of individuals.
- Sales – Sales management uses data mining to identify clusters, or groups of customers, who share similar patterns of buying behavior in order to implement different strategies for expanding their company’s overall market share with each audience. For example, such clusters might differentiate between customers who make infrequent, larger purchases and those who make more frequent yet smaller purchases.
Sample Analysis
Click on the images below to see a larger, full-size version in a separate window. 
Global customer view by officer
| 
Top customers matched to selected products
|
Resources
Demos Webinars White Papers/Tech Notes Brochures
|
Universal Business Intelligence 
