- Data Gathering
- Data Validation
- Data Preparation
- Feature Variable Calculation (creating more salient variables that are more predictive of target)
- Predictive Model Building and Testing
- Model Deployment
In our opinion, the main challenges are: Data Preparation, Feature Variable Calculation and Model Deployment. Probably 90% of time is spent on the above three areas.
Hello, I'm a DBA and developer interested in using models to try to help me out in predicting or understanding the database performance. There are views such as v$sesstat I'd like to sample every minute or so. Then I would classify them as database (or sessions) running OK or database (or sessions) facing issues. Also, I'd like to know if there is a specific statistic# set of statistics (features) that are outstanding in a way to help me out to understand what is going on. Would you mind giving me some directions of which data mining / Oracle data mining specifics docs I should read? You help is really appreciated.
Post a Comment