book

Artificial Intelligence and Machine Learning Fundamentals

by Zsolt Nagy

December 2018

Beginner

330 pages

5h 45m

English

Packt Publishing

Read now

Unlock full access

About the BookAbout the AuthorObjectivesAudienceApproachMinimum Hardware RequirementsSoftware RequirementsConventionsInstallation and SetupStarting AnacondaAdditional Resources
IntroductionHow does AI Solve Real World Problems?Diversity of DisciplinesFields and Applications of Artificial IntelligenceSimulating Intelligence – The Turing TestAI Tools and Learning ModelsClassification and PredictionLearning ModelsThe Role of Python in Artificial IntelligenceWhy is Python Dominant in Machine Learning, Data Science, and AI? Anaconda in PythonPython Libraries for Artificial IntelligenceA Brief Introduction to the NumPy LibraryExercise 1: Matrix Operations Using NumPyPython for Game AIIntelligent Agents in GamesBreadth First Search and Depth First SearchExploring the State Space of a GameExercise 2: Estimating the Number of Possible States in Tic-Tac-Toe GameExercise 3: Creating an AI RandomlyActivity 1: Generating All Possible Sequences of Steps in a Tic-Tac-Toe GameSummary
IntroductionExercise 4: Teaching the Agent to WinActivity 2: Teaching the Agent to Realize Situations When It Defends Against LossesActivity 3: Fixing the First and Second Moves of the AI to Make it InvincibleHeuristicsUninformed and Informed SearchCreating HeuristicsAdmissible and Non-Admissible HeuristicsHeuristic EvaluationExercise 5: Tic-Tac-Toe Static Evaluation with a Heuristic FunctionUsing Heuristics for an Informed SearchTypes of HeuristicsPathfinding with the A* AlgorithmExercise 6: Finding the Shortest Path to Reach a GoalExercise 7: Finding the Shortest Path Using BFSIntroducing the A* AlgorithmA* Search in Practice Using the simpleai LibraryGame AI with the Minmax Algorithm and Alpha-Beta Pruning Search Algorithms for Turn-Based Multiplayer GamesThe Minmax AlgorithmOptimizing the Minmax Algorithm with Alpha-Beta PruningDRYing up the Minmax Algorithm – The NegaMax AlgorithmUsing the EasyAI Library Activity 4: Connect FourSummary
IntroductionLinear Regression with One VariableWhat Is Regression?Features and LabelsFeature ScalingCross-Validation with Training and Test DataFitting a Model on Data with scikit-learn Linear Regression Using NumPy ArraysFitting a Model Using NumPy PolyfitPredicting Values with Linear RegressionActivity 5: Predicting PopulationLinear Regression with Multiple VariablesMultiple Linear RegressionThe Process of Linear RegressionImporting Data from Data SourcesLoading Stock Prices with Yahoo FinanceLoading Files with pandasLoading Stock Prices with QuandlExercise 8: Using Quandl to Load Stock PricesPreparing Data for PredictionPerforming and Validating Linear RegressionPredicting the FuturePolynomial and Support Vector RegressionPolynomial Regression with One VariableExercise 9: 1st, 2nd, and 3rd Degree Polynomial RegressionPolynomial Regression with Multiple VariablesSupport Vector RegressionSupport Vector Machines with a 3 Degree Polynomial KernelActivity 6: Stock Price Prediction with Quadratic and Cubic Linear Polynomial Regression with Multiple VariablesSummary
IntroductionThe Fundamentals of ClassificationExercise 10: Loading DatasetsData PreprocessingExercise 11: Pre-Processing DataMinmax Scaling of the Goal ColumnIdentifying Features and LabelsCross-Validation with scikit-learnActivity 7: Preparing Credit Data for ClassificationThe k-nearest neighbor ClassifierIntroducing the K-Nearest Neighbor AlgorithmDistance FunctionsExercise 12: Illustrating the K-nearest Neighbor Classifier AlgorithmExercise 13: k-nearest Neighbor Classification in scikit-learnExercise 14: Prediction with the k-nearest neighbors classifierParameterization of the k-nearest neighbor Classifier in scikit-learnActivity 8: Increasing the Accuracy of Credit ScoringClassification with Support Vector MachinesWhat are Support Vector Machine Classifiers?Understanding Support Vector MachinesSupport Vector Machines in scikit-learnParameters of the scikit-learn SVMActivity 9: Support Vector Machine Optimization in scikit-learnSummary
Introduction to Decision TreesEntropyExercise 15: Calculating the EntropyInformation GainGini ImpurityExit ConditionBuilding Decision Tree Classifiers using scikit-learnEvaluating the Performance of ClassifiersExercise 16: Precision and RecallExercise 17: Calculating the F1 ScoreConfusion MatrixExercise 18: Confusion MatrixActivity 10: Car Data ClassificationRandom Forest ClassifierConstructing a Random ForestRandom Forest Classification Using scikit-learnParameterization of the random forest classifierFeature ImportanceExtremely Randomized TreesActivity 11: Random Forest Classification for Your Car Rental CompanySummary
Introduction to ClusteringDefining the Clustering ProblemClustering ApproachesClustering Algorithms Supported by scikit-learnThe k-means AlgorithmExercise 19: k-means in scikit-learnParameterization of the k-means Algorithm in scikit-learnExercise 20: Retrieving the Center Points and the Labelsk-means Clustering of Sales DataActivity 12: k-means Clustering of Sales DataMean Shift AlgorithmExercise 21: Illustrating Mean Shift in 2DMean Shift Algorithm in scikit-learnImage Processing in PythonActivity 13: Shape Recognition with the Mean Shift AlgorithmSummary
Introduction TensorFlow for PythonInstalling TensorFlow in the Anaconda NavigatorTensorFlow OperationsExercise 22: Using Basic Operations and TensorFlow constantsPlaceholders and VariablesGlobal Variables InitializerIntroduction to Neural NetworksBiasesUse Cases for Artificial Neural NetworksActivation FunctionsExercise 23: Activation FunctionsForward and Backward PropagationConfiguring a Neural NetworkImporting the TensorFlow Digit DatasetModeling Features and LabelsTensorFlow Modeling for Multiple LabelsOptimizing the VariablesTraining the TensorFlow ModelUsing the Model for PredictionTesting the ModelRandomizing the Sample SizeActivity 14: Written Digit DetectionDeep LearningAdding LayersConvolutional Neural NetworksActivity 15: Written Digit Detection with Deep LearningSummary
Chapter 1: Principles of AIActivity 1: Generating All Possible Sequences of Steps in the tic-tac-toe GameChapter 2: AI with Search Techniques and GamesActivity 2: Teach the agent realize situations when it defends against lossesActivity 3: Fix the first and second moves of the AI to make it invincibleActivity 4: Connect FourChapter 3: RegressionActivity 5: Predicting PopulationActivity 6: Stock Price Prediction with Quadratic and Cubic Linear Polynomial Regression with Multiple VariablesChapter 4: ClassificationActivity 7: Preparing Credit Data for ClassificationActivity 8: Increase the accuracy of credit scoringActivity 9: Support Vector Machine Optimization in scikit-learnChapter 5: Using Trees for Predictive AnalysisActivity 10: Car Data ClassificationActivity 11: Random Forest Classification for your Car Rental CompanyChapter 6: ClusteringActivity 12: k-means Clustering of Sales Data Activity 13: Shape Recognition with the Mean Shift algorithmChapter 7: Deep Learning with Neural Networks Activity 14: Written digit detectionActivity 15 : Written Digit Detection with Deep Learning

Overview

"Artificial Intelligence and Machine Learning Fundamentals" is an accessible guide to understanding and applying the principles of AI and machine learning in Python. You will explore both foundational methods like regression and advanced techniques such as deep neural networks through step-by-step projects and hands-on exercises. This book empowers you to create real-life AI applications confidently.

What this Book will help me do

Understand essential AI concepts including search algorithms, regression, and classification.
Implement practical machine learning solutions using Python programming language.
Use decision trees and clustering methods for predictive data analysis.
Learn to design and train neural networks for advanced AI applications.
Apply theoretical AI concepts in real-world scenarios to solve problems effectively.

Author(s)

None Nagy is a seasoned professional in the field of artificial intelligence with years of experience in software development and data analysis. With a passion for teaching, Nagy specializes in breaking down complex concepts into digestible steps. This book reflects their practical approach for learners eager to delve into AI.

Who is it for?

This book is ideal for software developers and data scientists aiming to integrate AI or machine learning into their work. It's suitable for beginners as no prior AI knowledge is required. Having a background in Python programming and high school-level mathematics will be beneficial. Readers looking to explore AI applied to real-world challenges will find this book particularly valuable.