In this chapter, we learned how to define a business problem from a data science perspective through a well-defined, structured approach. We started by understanding how to approach a business problem, how to gather the requirements from stakeholders and business experts, and how to define the business problem by developing an initial hypothesis.
Once the business problem was defined with data pipelines and workflows, we looked into understanding how to start the analysis on the gathered data in order to generate the KPIs and carry out descriptive analytics to identify the key trends and patterns in the historical data through various visualization techniques.
We also learned how a data science project life cycle is structured, from defining the business problem to various pre-processing techniques and model development. In the next chapter, we will be learning how to implement the concept of high reproducibility on a Jupyter notebook, and its importance in development.