- - - - - - - - - - - -
KEY EXPECTED ACHIEVEMENTS
• The business need is understood and formalized in a descriptive datasheet or specifications
• The methods are clearly selected by their theoretical bases, advantages and drawbacks
• The data, its relevance and its source are described and prioritized
• The data are prepared (cleaned, enriched, mapped, agregated, …) and the approach is documented
• The data analysis is implemented and documented
• Machine Learning pipeline are built, tested and automatized for deployment
• The algorithms and models (descriptive and/or predictive) are developed and the approach is documented
• The results of the analysis or the models are presented to the customers with data visualisation tools indicating the performances and the limits. The source code is delivered and explained if necessary
• Models are validated and decisions deployment are taken, from roll-out-strategy
• Peer reviews are organized to ensure quality of delivering.