IL-FCT
Ilmek Tahmin
Hierarchical time-series forecasting for demand, sales, stock and energy load, with probabilistic ranges and automatic…
Forecasting & Recommendation View
Ilmek Öneri
Forecasting & Recommendation
IL-REC
Real-time personalised product, content and campaign recommendations for e-commerce, telecom and finance; churn prediction.
- Real-time personalisation
- Next best action
- Churn risk score
- A/B testing
Ilmek Öneri combines customer behaviour, product attributes and context signals to produce a ranking personalised for each user. The same model serves campaign, basket, e-mail and mobile-notification channels.
What it can do
- Real-time product recommendation and personalised search ranking
- Next best action: which offer to which customer through which channel
- Churn and payment-delay risk score
- A/B testing infrastructure and revenue-impact report
- Content-based recommendation for cold start; a session-based mode without personal data for privacy
For e-commerce customers average basket value has risen 8–14% and for telecom campaign conversion 20%.
How to start
Offline evaluation runs on historical basket and click data; then an A/B test runs on 10% of traffic. Revenue impact is reported with statistical significance before rolling out to all traffic.
| Latency | < 60 ms |
|---|---|
| Channels | Web, mobile, e-mail, SMS |
| Privacy mode | Session-based |
| Testing | A/B + revenue report |