Название: Modern Data Mining with Python: A risk-managed approach to developing and deploying explainable and efficient algorithms using ModelOps
Автор: Dushyant Singh Sengar, Vikash Chandra
Издательство: BPB Publications
Год: 2024
Страниц: 438
Язык: английский
Формат: epub (true)
Размер: 20.0 MB

"Modern Data Mining with Python" is a guidebook for responsibly implementing data mining techniques that involve collecting, storing, and analyzing large amounts of structured and unstructured data to extract useful insights and patterns.

Enter into the world of data mining and Machine Learning. Use insights from various data sources, from social media to credit card transactions. Master statistical tools, explore data trends, and patterns. Understand decision trees and artificial neural networks (ANNs). Manage high-dimensional data with dimensionality reduction. Explore binary classification with logistic regression. Spot concealed patterns with unsupervised learning. Analyze text with recurrent neural networks (RNNs) and visuals with convolutional neural networks (CNNs). Ensure model compliance with regulatory standards.

The book starts from the basics of statistics and exploratory data analysis and then ventures into advanced Deep Learning techniques. It emphasizes ethical Machine Learning model development, tackling biases, ensuring algorithmic transparency, and adhering to responsible AI principles. This approach is not only about learning techniques but also about becoming a responsible decision-maker in the data-driven business world.

Through its 13 meticulously crafted chapters, this book describes responsible AI approaches to improving various AI/ML techniques adoption and business processes by demystifying best practices in model risk management and operations. Author Dushyant Sengar created this guide for budding data scientists and experienced business leaders looking to improve real-world AI/ML system outcomes in the banking sector and beyond.

This book establishes a solid foundation of data mining, starting with exploratory data analysis and inferential statistics and progressing through advanced techniques like XGBoost, and deep learning. Each chapter, with a focus on a specific data mining technique and complete with its applications and nuances, serves as a building block of a comprehensive data mining knowledge base. You will explore industry best practices for making fair and efficient decisions while learning technical approaches for responsible AI across model validation, explainability, bias management, and AI-based product development using MLOps.

After reading this book, readers will be equipped with the skills and knowledge necessary to use Python for data mining and analysis in an industry set-up. They will be able to analyze and implement algorithms on large structured and unstructured datasets.

Contents: