Mage

Mage helped us with 1. The probability score for uptake of every product is calculated for customers using ML/ Regression models 2. Pick Top customers for a product/Top products for a customer, based on the requirement. 3. Identify popular product combinations using 4. Association rules from Market Basket Analysis (or affinity Analysis)\Bundle these products as combos 5. Alternatively, use fast-selling products as carriers to sell high-margin but low-selling products.

• Real-time monitoring.
• Alerting
• Advanced Drill Downs.

• Business Understanding.
• Data Acquisition and Understanding.
• Data Modeling and Evaluation.

We use Mage to setup ranking algorithms in our product wherever needed. We usually have multiple strategies for choosing the right temperature for air conditioners to function on. Depending on the variables inside and outside the room, we use Mage to rank the strategies and choose the best one.

• No-code tool.
• Easily integrate ranking algorithms.
• Cloud-based support.

• Reduced time to go live.
• Increased the conversion numbers.
• Improved accuracy of our AI models.

Amazon SageMaker, PyTorch on AWS and Cloudera Data Platform