Название: Artificial Intelligence for Automated Pricing Based on Product Descriptions
Автор: Nguyen Thi Ngoc Anh, Tran Ngoc Thang, Vijender Kumar Solanki
Издательство: Springer
Год: 2022
Страниц: 62
Язык: английский
Формат: pdf (true), epub
Размер: 11.3 MB

This book highlights Artificial Intelligence (AI) algorithms used in implementation of automated pricing. It presents the process for building automated pricing models from crawl data, preprocessed data to implement models, and their applications. The book also focuses on Machine Learning (ML) and Deep Learning (DL) methods for pricing, including from regression methods to hybrid and ensemble methods. The computational experiments are presented to illustrate the pricing processes and models.

Machine learning discovers new knowledge from big data collected via activities such as applications from mobile, social data, websites. Using machine learning help develop automatically adapt and customize the system by changing of data of individual users. Machine learning help the system recognizing that replace human such as faces, objects, handwriting characters, voice.

Machine learning is separated in three types: Supervised, unsupervised and reinforcement. Firstly, supervised learning includes classification and regression problems that output of classification is a discrete variable and output of regression is continuous variable. Secondly, Unsupervised learning is no desired output that learn something from data and latent relationships of data such as clustering that learns structure in the data. Finally, reinforcement learning defined an agent interacts with an environment and receives feedback reward signal.

Contents:
1. Pricing Based on Product Descriptions: Problem, Data and Methods
2. Machine Learning and Ensemble Methods
2. 1 Introduction to Machine Learning
2. 2 Linear Classification
2. 3 Support Vector Machine
2. 4 Decision Tree
2. 5 Ensemble Methods
3. Quantifying the Qualitative Features
4. Deep Learning Model for Product Classification
5. Product Feature Extraction from the Descriptions