ABSTRACT
A large number of data science and machine learning problems ultimately reduce to optimizations problems. New advents in AI, data science, machine learning, and deep learning affect almost all business and technology. This, however, doesn’t change the fact that human and natural resources are still finite and bounded by legal and ethical constraints. Advanced practices of machine learning may be helpful to guide scientists to achieve better solutions based on optimization at a faster clip, but they should keep on solving the same or rather complex problems as they were previously. A cascade of new data will support this process, but the expectations will rise as time goes on. Otherwise, what’s the point of AI or Big Data? For example, if a scientific team processes 2 GB of data and produces an optimal solution at a cost of Rs. 1000, the team should reduce the cost to Rs. 7000/- (instead of Rs. 10000/-) if it processes a “bigger” dataset of 20 GB.
In this chapter, the author discusses how to solve optimization problems using Python and other tools. The intention of the author is not to help the user become a skillful theoretician, but a skillful modeler. Therefore, little of mathematical principles related to the subject of optimization is discussed. Various aspects of optimization problems have been covered in the case studies mentioned in the chapter. The chapter can be effectively used to create easy yet powerful and efficient models.
