Abstract
Customer value is pushed as the first class information. The approach is to estimate directly the profit generated on a customer without estimating the conditional class probability. This methodology opens up new possibilities for profit estimation. In particular, association rules are used to summarize customer groups and to build a model for profit estimation.
| Original language | English |
|---|---|
| Pages | 738-740 |
| Number of pages | 3 |
| Publication status | Published - 2003 |
| Event | Nineteenth International Conference on Data Ingineering - Bangalore, India Duration: 5 Mar 2003 → 8 Mar 2003 |
Conference
| Conference | Nineteenth International Conference on Data Ingineering |
|---|---|
| Country/Territory | India |
| City | Bangalore |
| Period | 5/03/03 → 8/03/03 |