Abstract: "Being a well-studied economic problem, price optimization becomes increasingly complex in practice, especially when dealing with large enterprises holding significant market share. You have to account for complementary goods. If your competitor raises prices for cars and you sell gasoline, people will buy fewer cars and need less gasoline so it is a bad news for you. Alternatively, if your competitor raises prices for beef and you sell chicken, it is a good news for you as beef and chicken are substitute goods. As beef becomes more expensive, some people will switch to chicken.
Changes in prices among the goods you sell affect buyer’s decisions in the same way.
A good way of estimating demand in such a complex and inter-dependent environment is machine learning.
First, you need to collect the information about the items your competitors sell and their prices. Machine learning helps you identify those, which are similar to the items you sell.
Next, you train a model, which predicts how a change in prices for one good affects another.
After that, you build another machine learning model to predict the demand for every good.
Finally, you build an elasticity model, which takes into account additional business rules and generates price recommendations.
This talk is going to be about tips and pitfalls of designing such systems. We are going to discuss what be taken into account when training ML models for price recommendations, how to match different products, how to predict the demand and how to integrate the solution with the big data technologies."
Head of Data Science Office at Eleks