Artificial Intelligence

Artificial Intelligence Author: Djibril Chimère DIAW

Artificial Intelligence, published on April 2, 2023, is a comprehensive and systematically structured academic work that presents the theoretical foundations, historical development, algorithmic architectures, and applied domains of modern artificial intelligence. Conceived as both a reference volume and a structured educational resource, the book integrates machine learning, deep learning, natural language processing, computer vision, robotics, knowledge representation, optimization, evolutionary computation, and augmented intelligence into a unified conceptual framework.

The volume opens with a historical and epistemological overview of artificial intelligence, tracing its development from the foundational period of the 1950s and 1960s through the cycles of optimism and decline in the 1970s–1980s, the resurgence of the 1990s–2000s, and the contemporary era of large-scale data-driven AI systems. This contextual grounding situates modern advances within their intellectual lineage.

A substantial portion of the work is devoted to machine learning, including supervised, unsupervised, semi-supervised, and reinforcement learning. Regression models, classification algorithms, decision trees, support vector machines, ensemble methods, clustering techniques, dimensionality reduction, Q-learning, SARSA, and policy-based approaches are examined both conceptually and algorithmically. Deep learning architectures—convolutional neural networks (CNNs), recurrent neural networks (RNNs), LSTM, GRU, and generative adversarial networks (GANs)—are presented as extensions of statistical learning theory and representation learning.

The book provides extensive coverage of natural language processing (NLP), including sentiment analysis, named entity recognition, topic modeling, dependency parsing, summarization (extractive and abstractive), question answering systems, natural language generation, and machine translation. Both rule-based and statistical paradigms are addressed, alongside neural and transformer-based language models. Key algorithmic families such as Hidden Markov Models, Conditional Random Fields, word embeddings (Word2Vec, GloVe), transformer architectures, and large generative language models are analyzed within a coherent methodological structure.

Computer vision is treated in equal depth, covering traditional feature-based approaches and modern deep learning systems for object detection, image classification, image segmentation, and image captioning. Architectures such as RCNN variants, YOLO, U-Net, ResNet, DenseNet, and transformer-based vision models are integrated into a broader discussion of perception systems.

Beyond perception and language, the volume addresses robotics (kinematics, dynamics, control systems, motion planning, human–robot interaction), expert systems, cognitive computing, speech recognition, swarm intelligence algorithms, evolutionary computation, and optimization methods. Knowledge representation and reasoning are explored through propositional logic, first-order logic, description logic, modal and temporal logic, ontologies, semantic networks, constraint satisfaction, and automated planning.

The final sections extend the discussion toward augmented intelligence, human–computer interaction, intelligent agents, recommender systems, adaptive interfaces, and decision support systems, emphasizing the collaborative dimension between human reasoning and computational intelligence.

Rather than presenting artificial intelligence as a collection of isolated techniques, this book articulates AI as an integrated scientific discipline grounded in mathematics, algorithmic reasoning, formal logic, optimization theory, and computational systems engineering. The structure progresses from foundational concepts to advanced architectures and interdisciplinary applications, making the work suitable for students, independent researchers, engineers, and professionals seeking a rigorous and comprehensive overview.

This first edition captures a particular moment in the evolution of artificial intelligence research and practice. Future editions may expand or refine specific sections as the field continues to advance. Nevertheless, the present volume is conceived as a durable intellectual contribution, preserving a structured and systematic vision of AI in the early decades of the twenty-first century.

Edition information

Publié le : 2 avril 2023
First published: [2023/04/02]
Current version: v1.0
Last updated: [2023/04/02]

Author: [Djibril Chimère DIAW]
Original language: [ENGLISH]
Digital publication: Archive.org
Collection: [Artificial Intelligence]


https://archive.org/details/artificial-intelligence_202602





Contents
Artificial Intelligence    1
Copyright    2
Author’s Note    3
About The Author    4
Dedication    6
To all mothers,    7
Artificial Intelligence (AI)    15
Milestones in AI    16
1950s-1960s :The birth of AI    16
1950s-1960s Perspectives    17
1970s-1980s : The rise and fall of AI    18
1990s-2000s : The resurgence of AI    19
2010s-present :The age of AI    20
1 Machine Learning (ML)    21
1.1 Supervised Learning    22
1.1.1 Regression    23
1.1.1.1 Linear regression    24
1.1.1.2 Polynomial regression    25
1.1.1.3 Ridge regression    26
1.1.1.4 Lasso regression    27
1.1.1.5 Support vector regression (SVR)    28
1.1.2 Classification    29
1.1.2.1 Logistic regression    30
1.1.2.2 Decision trees    31
1.1.2.3 Random forests    32
1.1.2.4 Support vector machines    33
1.1.2.5 Neural networks    34
1.2 Unsupervised Learning    35
1.2.1 Clustering    36
1.2.1.1 K-means clustering    37
1.2.1.2 Hierarchical clustering    38
1.2.1.3 Density-based clustering    39
1.2.2 Dimensionality reduction    40
1.2.2.1 Feature selection    41
1.2.2.2 Feature extraction    42
1.3 Semi-supervised Learning    43
1.3.1 Self-training    44
1.3.2 Co-training    45
1.3.3 Graph-based methods    46
1.4 Reinforcement Learning    47
1.4.1 Value-based methods    48
1.4.1.1 Q-learning    49
1.4.1.2 SARSA (State-Action-Reward-State-Action)    50
1.4.2 Policy-based methods    51
1.4.2.1 Policy gradient method    52
1.4.2.2 Actor-critic method    53
2 Deep Learning    54
2.1 Convolutional Neural Networks (CNN)    55
2.2 Recurrent Neural Networks (RNN)    56
2.2.1 Simple Recurrent Neural Networks    57
2.2.2 Gated Recurrent Units (GRU) Networks    58
2.2.3 Long Short-Term Memory (LSTM) networks    59
2.3 Generative Adversarial Networks (GAN)    60
3 Natural Language Processing (NLP)    61
3.1 Sentiment Analysis    62
3.1.1 Rule-based methods    63
3.1.2 Machine learning-based methods    64
3.2 Named Entity Recognition (NER)    65
3.3 Text Summarization    66
3.3.1 TextRank    67
3.3.2 Latent Semantic Analysis (LSA)    68
3.3.3 Latent Dirichlet Allocation (LDA)    69
3.4 Language Translation    70
3.4.1 Rule-based machine translation    71
3.4.2 Statistical machine translation    72
3.4.3 Neural machine translation    73
4 Computer Vision (CV)    74
4.1 Object Detection    75
4.1.1 Traditional computer vision methods    76
4.1.1.1 Feature detection and extraction    77
4.1.1.2 Template matching    78
4.1.1.3 Image segmentation    79
4.1.1.4 Optical flow    80
4.1.1.5 Hough transform    81
4.1.2 Region-based convolutional neural networks (RCNNs)    82
4.1.2.1 Fast R-CNN    83
4.1.2.2 Faster R-CNN    84
4.1.2.3 Mask R-CNN    85
4.1.3 Single-shot detectors    86
4.1.3.1 VGG (Visual Geometry Group)    87
4.1.3.2 ResNet (Residual Network)    88
4.1.4 You Only Look Once (YOLO)    89
4.1.4.1 Anchor boxes    90
4.1.4.2 Non-maximum suppression    91
4.2 Image Segmentation    92
4.2.1 Thresholding    93
4.2.2 Edge-based segmentation    94
4.2.3 Region-based segmentation    95
4.2.4 Watershed segmentation    96
4.2.5 Deep learning-based segmentation    97
4.2.5.1 U-Net    98
4.2.5.2 Fully Convolutional Networks (FCN)    99
4.2.5.3 DeepLab    100
4.2.5.4 SegNet    101
4.3 Image Classification    102
4.3.1 Traditional computer vision methods    103
4.3.1.1 Bag-of-Visual-Words model    104
4.3.1.2 Histogram of Oriented Gradients (HOG) feature descriptor    105
4.3.2 Deep learning-based methods    106
4.3.2.1 Convolutional neural networks (CNNs)    107
4.3.2.2 Transfer learning    108
4.3.2.3 Residual networks (ResNets)    109
4.3.2.4 Dense convolutional networks (DenseNets)    110
4.3.2.5 Inception networks    111
4.4 Image Captioning    112
4.4.1 CNN-RNN model    113
4.4.2 Encoder-Decoder model    114
4.4.3 Attention-based model    115
4.4.4 Transformer-based model    116
5 Robotics    117
5.1 Kinematics    118
5.1.1 Forward kinematic    119
5.1.2 Inverse kinematics    120
5.2 Dynamics    121
5.2.1 Forward Dynamics    122
5.2.2 Inverse Dynamics    123
5.3 Control Systems    124
5.3.1 Classical control theory    125
5.3.2 Fuzzy logic control    126
5.3.3 Neural network-based control    127
5.4 Robot perception    128
5.5 Motion planning    129
5.6 Human-robot interaction    130
6. Expert Systems    131
6.1 Rule-based Systems    132
6.2 Knowledge-based Systems    133
6.3 Decision Support Systems    134
6.4 Fuzzy Logic Systems    135
7. Cognitive Computing    136
7.1 Speech Recognition    137
7.1.1 Acoustic modeling    138
7.1.1.1 Hidden Markov Models    139
7.1.1.2 Gaussian Mixture Models (GMMs)    140
7.1.1.3 Deep Neural Networks (DNNs)    141
7.1.1.4 Convolutional Neural Networks (CNNs)    142
7.1.1.5 Recurrent Neural Networks (RNNs)    143
7.1.2 Language modeling algorithms    144
7.1.2.1 n-gram models    145
7.1.2.2 Neural language models    146
7.1.2.2.1 Recurrent neural network language model (RNNLM)    147
7.1.2.2.2 Transformer language model    148
7.1.2.3 Transformer-based models    149
7.1.2.4 Maximum Entropy Markov Models    150
7.1.2.5 Conditional Random Fields    151
7.1.2.6 Probabilistic context-free grammars    152
7.1.2.7 Word embeddings    153
7.1.2.7.1 Word2Vec    154
7.1.2.7.2 GloVe    155
7.1.2.8 Recurrent neural networks (RNNs)    156
7.1.2.9 Hidden Markov Models (HMMs)    157
7.1.3 Decoding algorithms    158
7.1.3.1 Viterbi algorithm    159
7.1.3.2 Beam search    160
7.1.3.3 CYK algorithm    161
7.1.3.5 Earley algorithm    162
7.2 Emotion Detection    163
7.2.1 Support Vector Machines (SVMs)    165
7.2.2 Random Forest    166
7.2.3 Deep Learning applications    167
7.2.4 Naive Bayes applications    168
7.2.5 K-Nearest Neighbors (KNN)    169
7.2.6 Decision Trees applications    170
7.3 Natural Language Understanding    171
7.3.1 Information Retrieval algorithm    172
7.3.1.1 Boolean retrieval    174
7.3.1.2 Vector space model (VSM)    175
7.3.1.3 Latent Semantic Analysis (LSA)    176
7.3.1.4 Latent Dirichlet Allocation (LDA)    177
7.3.1.5 PageRank    178
7.3.1.6 TF-IDF (Term Frequency-Inverse Document Frequency)    179
7.3.2 Topic Modeling algorithm    180
7.3.2.1 Non-negative Matrix Factorization (NMF)    181
7.3.2.2 Hierarchical Dirichlet Process (HDP)    182
7.3.2.3 Latent Semantic Analysis (LSA)    183
7.3.2.4 Latent Dirichlet Allocation (LDA)    184
7.3.3 Dependency Parsing    185
7.3.3.1 Arc-standard parsing algorithm    186
7.3.3.2 Arc-eager parsing algorithm    187
7.3.4 Summarization algorithm    188
7.3.4.1 Extractive summarization    189
7.3.4.1.1 TextRank    190
7.3.4.1.2 LexRank    191
7.3.4.1.3 Latent Semantic Analysis (LSA)    192
7.3.4.2 Abstractive summarization    193
 7.3.4.2.1 Pointer-Generator Networks    194
7.3.4.2.2 Transformer models    195
7.3.4.2.2.1 Bidirectional Encoder Representations from Transformers,    196
7.3.4.2.2.2 GPT-2 (Generative Pre-trained Transformer 2)    197
7.3.4.2.3 GPT-3    198
7.3.5 Question Answering (QA) algorithm    199
7.3.5.1 Information Retrieval-Based QA    200
7.3.5.2 Knowledge-Based QA    201
7.3.5.3 Language Model-Based QA    202
7.3.5.4 Hybrid QA    203
7.3.6 Natural Language Generation algorithms    204
7.3.6.1 Template-based generation    205
7.3.6.2 Rule-based generation    206
7.3.6.3 Markov Chain-based generation    207
7.3.6.4 Neural network-based generation    208
7.3.6.5 GPT (Generative Pre-trained Transformer)    209
7.3.6.6 Text-to-Speech (TTS) generation    210
7.3.6.7 Planning-based generation    211
7.3.6.8 Evolutionary algorithms-based generation    212
7.3.7 Machine Translation algorithm    213
7.3.7.1 Rule-based Machine Translation    214
7.3.7.2 Statistical Machine Translation    215
7.3.7.3 Neural Machine Translation    216
7.3.7.4 Hybrid Machine Translation    217
7.3.7.5 Example-based Machine Translation    218
7.3.7.6 Transfer-based Machine Translation    219
7.3.7.7 Interactive Machine Translation    220
7.3.8 Named Entity Recognition (NER)    221
7.3.8.1 Rule-Based NER    222
7.3.8.2 Statistical NER    223
7.3.8.3 Machine Learning NER    224
7.3.8.4 Deep Learning NER    225
7.3.8.5 Hybrid NER    226
7.3.8.6 Unsupervised NER    227
8 Swarm Intelligence    228
8.1 Ant Colony Optimization    229
8.2 Artificial Bee Colony    231
8.3 Bat Algorithm    232
8.4 Bacterial Foraging Optimization    233
8.5 Cuckoo Search (CS) algorithm    234
8.6 Firefly Algorithm    235
8.7 Fish Swarm Optimization    236
8.8 Genetic Algorithm    237
8.9 Glowworm Swarm Optimization    239
8.10 Grey Wolf Optimizer    240
8.11 Harmony Search    241
8.12 Particle Swarm Optimization    242
8.13 Social Spider Optimization    243
8.14 Whale Optimization Algorithm    244
9 Knowledge Representation and Reasoning algorithms    245
9.1 Logic-based approaches    247
9.1.1 Propositional logic    248
9.1.2 First-order logic    249
9.1.3 Description logic    250
9.1.4 Modal logic    251
9.1.4.1 Propositional modal logic    252
9.1.4.2 First-order modal logic    253
9.1.4.3 Dynamic logic    254
9.1.5 Default Logic    255
9.1.6 Fuzzy Logic    256
9.1.7 Temporal logic    257
9.2 Ontologies    258
9.3 Semantic Networks    260
9.3.1 Conceptual graph    261
9.3.2 Ontology engineering    263
9.3.3 Semantic Web    265
9.4 Frames    267
9.5 Case-Based Reasoning    269
9.6 Inductive Logic Programming    270
9.7 Constraint Satisfaction    271
9.7.1 Backtracking    272
9.7.2 Constraint propagation    273
9.7.2.1 Forward checking    275
9.7.2.2 Arc consistency    276
9.7.3 Local search    277
10 Planning and Scheduling    278
10.1 Automated Planning    280
10.2 Temporal Reasoning    281
10.3 Resource Allocation    282
10.4 Activity Recognition    283
10.5 Planning    284
10.5.1 Forward planning    287
10.5.2 Backward planning    288
10.5.3 State-space search    289
10.5.4 Heuristic search    290
10.6 Scheduling    291
10.6.1 Constraint-based scheduling    292
10.6.2 Resource-constrained scheduling    293
10.6.3 Dynamic scheduling    294
10.6.4 Optimization-based scheduling    295
11 Evolutionary Computation    296
11.1 Genetic Algorithms    297
11.2 Genetic Programming    298
11.3 Evolution Strategies    299
11.4 Evolutionary Programming    300
11.5 Differential Evolution    301
11.6 Artificial Immune Systems    302
11.6.1 Clonal selection algorithm (CSA)    303
11.6.2 Artificial immune network    304
11.6.1 Clonal selection algorithm (CSA)    305
11.6.2 Artificial immune network    306
 11.6.3 Negative selection algorithm    307
11.6.3 Negative selection algorithm    308
12. Augmented Intelligence    309
12.1 Human-Computer Interaction    310
12.2 Decision Support Systems    311
12.2.1 Model-driven Decision Support Systems    312
12.2.2 Data-driven Decision Support Systems    314
12.2.3 Knowledge-driven Decision Support Systems    316
12.3 Intelligent User Interfaces    318
12.3.1 Adaptive Interfaces    320
12.3.2 Intelligent Agents    321
12.3.3 Multimodal Interfaces    323
12.3.4 Augmented Reality Interfaces    324
12.3.5 Conversational Interfaces    325
12.4 Recommender Systems    326
12.4.1 Content-based recommender systems    327
12.4.2 Collaborative filtering recommender systems    328
12.4.3 Hybrid recommender systems    330
Biliography    331
Books    331
Articles    334