Engineering: Cloud and data engineering forms the foundation of the architectural design, enabling enhanced efficiency, scalability, and insights, and comprises the following components:
• Infrastructure setup
To establish a scalable computational ecosystem, we leveraged cloud-native services like BigQuery, Compute Engine, Cloud Storage, GCP Marketplace (Neo4J), and Google AppSheet. Equivalent services are available on other popular cloud service providers such as AWS and Azure. For data processing, we utilized Python libraries like Pandas and NumPy, enabling both standard and customized data processing capabilities.
• Data harmonization
We consolidated more than 10 disparate data sources, encompassing past purchases, logins, streaming sessions, demographics, and registrations. Through this process, we established a single source of truth known as Customer 360 . This unified data repository enables us to generate a comprehensive user context, empowering us with deep insights into user behavior, preferences, and engagement patterns.
• Feature engineering
We employed advanced data transformation and feature creation techniques to curate an enhanced Customer 360, comprising over 500 features . These features are the foundation for quantifying customer behavior across multiple dimensions and capturing diverse latent interactions.
© 2023 Fractal Analytics Inc. All rights reserved
04
Made with FlippingBook - PDF hosting