Practical Machine Learning Fundamentals (Video Course)

Description

• Copyright 2027
• Edition: 1st

• Online Video
• ISBN-10: 0-13-592786-2
• ISBN-13: 978-0-13-592786-1

12+ Hours of Video Instruction

Build a strong foundation in machine learning through approachable explanations, real-world examples, and hands-on Python labs.

Overview:

Practical Machine Learning Fundamentals guides you through the complete machine learning workflow, from exploring and preparing data to building, evaluating, securing, and explaining models with tools such as Jupyter, pandas, and scikit-learn. Youll develop practical experience with preprocessing pipelines, classification models, performance metrics, and responsible machine learning, while learning to recognize security threats, bias, and other common risks. Designed for learners who are new to machine learning, this course provides the essential knowledge and practical skills needed to begin creating trustworthy ML solutions and continue toward more advanced study.

Topics Include:

• Machine learning concepts, major learning types, real-world applications, and the end-to-end ML workflow
• Python-based data exploration and preparation using Jupyter, Google Colab, pandas, pyTorch, and scikit-learn
• Classification models, train/test splits, cross-validation, overfitting, and model evaluation metrics
• Machine learning security threats, adversarial attacks, threat modeling, monitoring, and secure deployment considerations
• Responsible machine learning, including bias, fairness metrics, explainability tools, and career pathways
• Feature engineering and preprocessing: missing data, outliers, categorical encoding, and scaling

Skill Level:

• Beginner to Intermediate

Learn How To:

• Distinguish machine learning from artificial intelligence and data science, and select an appropriate learning approach for a given problem.
• Explore, clean, transform, and prepare real-world datasets using Python, pandas, and scikit-learn pipelines.
• Build and evaluate classification models using cross-validation, confusion matrices, precision, recall, F1 scores, and ROC and precision-recall curves.
• Identify adversarial attacks and other ML security risks, and apply foundational practices for protecting data, models, and deployments.
• Assess model bias, apply fairness metrics, and explain predictions using SHAP, LIME, permutation importance, and partial dependence plots.

Course Requirement:

• Basic familiarity with Python, including variables, functions, importing, and using libraries such as NumPy
• No previous machine learning experience required
• Introductory awareness of cybersecurity concepts (such as the confidentiality, integrity, and availability triad) is helpful but not required

Who Should Take This Course:

• Software developers, data analysts, and engineers exploring machine learning for the first time
• Cybersecurity professionals interested in understanding the security implications of machine learning
• Product and project managers responsible for overseeing machine learning initiatives
• Students preparing for advanced machine learning study or ML certification pathways

About Pearson Video Training:   

Pearson publishes expert-led video tutorials covering a wide selection of technology topics designed to teach you the skills you need to succeed. These professional and personal technology videos feature world-leading author instructors published by your trusted technology brands: Addison-Wesley, Cisco Press, Pearson IT Certification, Prentice Hall, Sams, and Que Topics include: IT Certification, Network Security, Cisco Technology, Programming, Web Development, Mobile Development, and more.  Learn more about Pearson Video training at  http://www.informit.com/video.



Sample Content

Table of Contents

Introduction

Module 1:          ML Foundations & Core Concepts

Lesson 1:            What is Machine Learning?

1.1          ML vs. AI vs. data science

1.2          Why ML matters in industry

1.3          Security context: Prediction as attack surface

Lesson 2:            Major ML Types

2.1          Supervised, unsupervised, reinforcement learning

2.2          Use cases across domains

2.3          When to choose each

Lesson 3:            Real-World ML Systems

3.1          Recommendation engines

3.2          Fraud detection

3.3          Predictive maintenance

3.4          Cybersecurity ML workflows

Lesson 4:            ML Workflow Overview

4.1          The end-to-end ML workflow

4.2          Common failures & misunderstandings

Module 1 Lab: Exploring ML in Python

Module 2:          Python ML Tooling & Data Foundations

Lesson 5:            Jupyter & Google Colab Deep Dive

5.1          File management

5.2          GPU/CPU runtimes

5.3          Notebook best practices

Lesson 6:            pandas for Data Analysis

6.1          Selecting, filtering, grouping

6.2          Handling missing values

6.3          Joining datasets

Lesson 7:            Understanding Real-World Data

7.1          Data types

7.2          Leakage and why it matters

7.3          Schema validation basics

Module 2 Lab: Exploring ML in Python

Module 3:          Feature Engineering & Preprocessing

Lesson 8:            Missing Data Handling

8.1          Mean/median/mode

8.2          Domain-driven imputations

8.3          When not to impute

Lesson 9:            Outliers & Robust Preprocessing

9.1          Z-score

9.2          IQR

9.3          Winsorization

9.4          Isolation forest

9.5          Why multivariate outliers matter

Lesson 10:        Encoding Categorical Data

10.1       Label encoding

10.2       One-hot

10.3       Target encoding

Lesson 11:        Scaling & Normalization

11.1       StandardScaler and MinMaxScaler

11.2       When scaling matters

11.3       Pipeline integration

Module 3 Lab: Titanic Preprocessing Pipeline

Module 4:          Modeling & Evaluation

Lesson 12:        Classification Foundations

12.1       Logistic regression intuition

12.2       Decision boundaries

12.3       Overfitting vs. underfitting

Lesson 13:        Train/Test Split & Cross-Validation

13.1       Why we split data

13.2       Stratified sampling

13.3       K-fold and repeated Cross-Validation (CV)

13.4       Time-series CV overview

Lesson 14:        Model Evaluation Metrics

14.1       Accuracy pitfalls

14.2       Precision, recall, and F1

14.3       Confusion matrix

14.4       Receiver Operating Characteristic (ROC) vs. Precision-Recall Curves

14.5       Business context of metrics

Module 4 Lab: Build & Evaluate a Titanic Classifier

Module 5:          ML Security & Adversarial Thinking

Lesson 15:        ML Threat Landscape

15.1       Data poisoning, label flipping, and model inversion

15.2       Membership inference

15.3       Adversarial examples

Lesson 16:        Attack Demonstration

16.1       Fast Gradient Sign Method (FGSM)

16.2       Black-box adversarial inputs

16.3       Model extraction

Lesson 17:        Defending ML Systems

17.1       Threat modeling

17.2       Data validation and versioning

17.3       Monitoring drift and abuse

17.4       Secure deployment considerations

Module 5 Lab: Crafting & Detecting Adversarial Attacks

Module 6:          Responsible ML, Fairness & Career Next Steps

Lesson 18:        Understanding ML Bias

18.1       Sources of bias

18.2       Examples from healthcare, justice, and finance

18.3       Why fairness is hard

Lesson 19:        Fairness Metrics

19.1       Demographic parity

19.2       Equalized odds

19.3       Individual fairness

19.4       Counterfactual fairness

Lesson 20:        Explainability Tools

20.1       SHapley Additive exPlanations (SHAP)

20.2       Local Interpretable Model-agnostic Explanations (LIME)

20.3       Permutation importance

20.4       Partial dependence plots

Lesson 21:        Career Pathways & Learning Roadmap

21.1       Data science vs. ML engineering vs. MLOps

21.2       Building a project portfolio

21.3       Certification choices

Module 6 Lab: Fairness Evaluation on the Titanic Model



Updates

Submit Errata

