Data aggregation consolidates granular records into summarized, meaningful measures. For user segmentation, individual clicks and other events become user-level behaviors that a model can compare. section 4.1.2
Useful aggregated behaviors include:
- Purchased items and total purchase value. section 4.1.2
- Recommendation clicks and website visits. section 4.1.2
- Add-to-cart and checkout frequency. section 4.1.2
From events to segmentation features
Instead of analyzing every click, summarize each user’s purchases, purchase value, recommendation use, visits, and shopping-flow engagement. These measures can reveal engagement, value, or possible churn risk. section 4.1.2
Table 4.1 Examples of data points for clustering
| Features |
|||||
| user_id |
purchased_items |
purchase_value |
last_active |
n_visits |
|
| Examples |
abhj3k |
2 |
908 |
2024-04-30 08:36:24 |
48 |
| shj67d |
0 |
0 |
2023-12-26 12:56:24 |
24 |
|
| i963gh |
12 |
673 |
2024-05-15 23:22:11 |
156 |
|
Compare aggregated user features such as purchases, purchase value, last activity, and visit count. section 4.1.2
