Operational Data Science: Part 2

Data science projects are often depicted as following round circle methodology.  suggesting that it repeats itself forever.  We provided two examples in Oper atio nal  Data  Sci ence : Pa rt 1 .  Truth is that lots of data science projects run the process only once.  They star t wi th a  bus ines s qu esti on from a business stakeholder. They then go on with data understanding, data preparation, data modeling, model evaluation, and reporting.  They stop once results are reported back to the skakeholder who asked the question.  They would be better depicted as a linear process: There are many great examples of this linear process.   Lucius Riccio’s Pothole Analytics is a good example.  Riccio did a great job at answering a business question, namely “how to better repair potholes on NYC streets?” …


Link to Full Article: Operational Data Science: Part 2