While the volume of enterprise data is growing exponentially,
the capacity to analyze the data has remained relatively
unchanged. Genalytics has developed its Model product
to break the analytic bottleneck, reduce cycle times
and develop predictive models that significantly outperform
all others. The key benefits of our predictive models
include:
- 10% better predictive accuracy
- Reduced model development time by 50-75% through
automated data attributes
- Exploration of hundreds of data attributes to provide
better insight
- Elimination of manual data preparation
- Updated models on a daily or weekly basis to reflect
the most recent customer data
Applications
- Direct marketing
- Credit scoring
- Risk management
- Fraud detection
- Cross-sell/up-sell promotion
|
System Requirements
- Relational database with a JDBC driver supporting JDBC2.0 (such as Oracle, DB2, Informix, SQL Server, Interbase, MySQL)
- Minimum configuration: Three servers, each with a dual CPU Pentium 4 or similar processor running at 3+ GHz, 1GB of RAM and at least 40 GB of storage
- 100 MB network is recommended to connect multiple servers
|
|
|
|
|
|
Model Architecture
Genalytics Model runs on a 100% Java platform that is easy to install
on Linux, Solaris or Windows. The system is scalable and distributed,
and can run as a standalone or embedded application within your
existing CRM or Data Warehousing infrastructure. A development API
is also available.
Model also provides finished models in SAS or XML code
for integration with your current operations.
You can install the software on one of multiple servers,
and the user interface can run on any Java-enabled workstation
on the network. Additional Fitness Evaluators can be
added to increase the overall performance. |
|