Development of a machine learning model for predicting product conversions on the Wildberries marketplace

This study recearching the use of a machine learning model to predict conversion rates on the Wildberries marketplace. Due to the fact that conversion is a value whose statistical significance is highly depend on the product's lifespan, this paper helps partially solve the cold start problem and improve the quality of search results. There were also explored several approaches to creating a dataset and training a model. A pipeline was also written for regularly uploading data about a certain subset of search sessions and automatically loading them into an inner database. The main focus of this research was on applying the knowledge in the field of applied mathematics and computer science to an issue relevant to business development. As a part of this work, various aspects of machine learning were studied and applied in the search ranking field.