Segmentation of users using machine learning methods
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
В данной дипломной работе исследуется задача сегментации пользователей для эффективного взаимодействия с рынком. В рамках реализации был произведен обзор существующих подходов и практик, применяемых в условиях реальных бизнес-процессов. Проведен анализ данных, а также выбраны необходимые методы для препроцессинга, анализа и уменьшения размерности данных. Рассмотрены математические модели, применимые к поставленной задаче, их преимущества и недостатки. Произведено сравнение построенных моделей. На основе данных интернет - сервиса для размещения объявлений была предложена сегментация пользователей на базе модели машинного обучения и внедрено программное решение, автоматизирующее процесс передачи клиентов между менеджерами.
The graduation paper examines the problem of segmentation of users for effective interaction with the market. As part of the implementation, an overview of existing approaches and practices applied in real business processes was carried out. Data analysis was carried out, and the necessary methods for preprocessing, analysis and data dimensionality reduction were selected. Mathematical models applicable to the task, their advantages and disadvantages are considered. The comparison of the constructed models is made. Based on the data of the Internet service for placing ads, segmentation of users based on a machine learning model was proposed and a software solution was implemented that automates the process of transferring clients between managers.
The graduation paper examines the problem of segmentation of users for effective interaction with the market. As part of the implementation, an overview of existing approaches and practices applied in real business processes was carried out. Data analysis was carried out, and the necessary methods for preprocessing, analysis and data dimensionality reduction were selected. Mathematical models applicable to the task, their advantages and disadvantages are considered. The comparison of the constructed models is made. Based on the data of the Internet service for placing ads, segmentation of users based on a machine learning model was proposed and a software solution was implemented that automates the process of transferring clients between managers.