Data Mining Flashcards
What are the 6 stages in the data mining step diagram?
- Business Understanding: Determine objectives, plan.
- Data Understanding: Collect the data, verify data quality.
- Data Preparation: Clean/Construct/Format data.
- Modelling: Select modelling technique, build model, assess model.
- Evaluation: Evaluate results, review process, determine next steps.
- Deployment: Plan deployment, produce final report, review project.
Can you describe for me 4 methods of data mining?
Clustering: Group data together based on similarity.
Prediction: Data produced is based on past events and current trends, for example assuming a customer with bad credit history won’t take credit risk.
Outlier detection: Specifically identifying an anomaly in a data set. For example, why did our usually mostly Male customer base see an increase in females during summer? You would investigate this anomaly.
Tracking patterns: Literally just noticing patterns in data sets. Like how your website gets a spike in visitors on hot days then attempting to capitalise on this.