Intelligent natural language processing

Preface......Page 6
Contents......Page 8
Sentiment Analysis......Page 12
 1 Introduction......Page 13
 2 Background......Page 15
 3 Data Collection and Annotation......Page 16
 4 Building Word Embeddings......Page 17
 5 The Implemented Models......Page 18
  5.1 The Baseline Model......Page 19
   5.2.1 Bidirectional Long Short-Term Memory Networks (BI-LSTMs)......Page 20
   5.2.3 BI-LSTM-CRF......Page 21
 6 Performance Evaluation......Page 23
 References......Page 24
 Abstract......Page 26
 1 Introduction......Page 27
 2 Related Work......Page 28
  3.1 Encoding......Page 29
  3.3 Excess Resources Required......Page 30
  3.4 Sarcastic Tamper......Page 31
  3.5 One Word Represents Two Polarities......Page 32
  3.6 Indifferent Writing Style......Page 33
  3.8 Word Short Forms......Page 34
  3.9 Same Word Usage for Both Polarities......Page 35
 5 Implementation of Arabic Sentiment Analysis......Page 36
 6 Evaluation and Results......Page 40
 References......Page 41
 1 Introduction......Page 44
 2 Related Work......Page 46
 3 Arabic Sentiment Analysis......Page 48
  4.2 Properties......Page 49
  5.1 Text Pre-processing......Page 51
 6 Experimental Results......Page 53
  6.1 Bag of Words......Page 54
  6.2 Lexicon-Based Classification......Page 57
 7 Conclusions......Page 59
 References......Page 60
 1 Introduction......Page 62
  3.1 Corpus Collection and Preparation......Page 64
  3.2 Pre-processing......Page 65
  3.3 Text Classification......Page 69
 4 Results and Evaluation......Page 71
 5 Conclusion and Future Work......Page 73
 References......Page 74
 1 Introduction......Page 76
 2 Related Research......Page 79
 4 Experimental Set-Up......Page 81
  4.1 Data Sets......Page 82
  4.2 Features......Page 84
  4.3 Training and Test Regimes......Page 90
  4.4 Evaluation and Baselines......Page 93
  4.5 Hyperparameter Optimization......Page 94
 5 Test Results......Page 97
 6 Discussion......Page 98
 7 Conclusions and Future Work......Page 103
 References......Page 104
Machine Translation......Page 107
 1 Introduction......Page 108
  2.1 Modern Standard Arabic Challenges for MT......Page 110
  2.2 Arabic Language in Microblogs......Page 112
 3 Overview of Statistical Machine Translation......Page 113
  3.2 Word Alignment......Page 115
  3.4 Decoding......Page 117
  4.2 Data Collection......Page 118
  4.4 Results......Page 121
  4.5 Discussion......Page 122
 5 Conclusion and Future Work......Page 123
 References......Page 124
Developing a Transfer-Based System  for Arabic Dialects Translation......Page 127
 1 Introduction......Page 128
 2 Related Studies......Page 129
 3 Arabic Language Variation......Page 131
  4.2 Statistical MT......Page 132
 5 (ALMoFseH) Arabic Dialects Machine Translation......Page 133
  6.1 Building a Lexical Database......Page 134
  6.2 The Transfer System......Page 136
  6.3 Naive Bayesian Classifier (NB)......Page 137
  6.4 Rewrite Rules for Dialectal Normalization......Page 138
 7 Evaluation of the System......Page 140
 8 Results......Page 142
 References......Page 143
 1 Introduction......Page 145
  2.1 Classical Arabic......Page 147
  2.2 Modern Standard Arabic......Page 148
  3.1 Metaphor Translation......Page 149
  3.2 Metaphor in Holy Quran......Page 153
  3.4 Metaphor in Dialect Arabic......Page 155
 4 Named Entity Recognition Translation......Page 156
 5 Word Sense Disambiguation Translation......Page 158
 6 Conclusion......Page 160
 References......Page 161
Information Extraction......Page 163
Graph-Based Keyword Extraction......Page 164
 1 Introduction......Page 165
 2 Related Work......Page 166
  3.1 Overview......Page 167
  3.2 The Proposed Methodology......Page 168
  3.3 Dataset......Page 170
  3.4 Data Processing......Page 171
  3.5 Learn Classifier and TF/IDF......Page 172
  3.6 Performance Evaluation......Page 174
 5 Conclusion and Future Prospects......Page 176
 References......Page 177
 Abstract......Page 178
 1 Introduction......Page 179
  2.1 Previous NE Categorization......Page 180
  2.2 NER Approaches and Systems......Page 181
  3.1 Identification of the ANE Forms and Categories......Page 184
  3.2 Relationship Between ANEs......Page 188
 4 Proposed Method......Page 189
  4.1 Analysis Transducer Establishment......Page 190
  4.2 Synthesis Transducer Establishment......Page 194
 5 Implementation and Evaluation......Page 195
 6 Conclusion......Page 200
 References......Page 201
 Abstract......Page 204
 1 Introduction......Page 205
 2 Arabic Semantic Relation Extraction and Ontology Learning......Page 206
 3 Arabic Semantic Relation Extraction......Page 207
   3.1.1 Rule-Based Approach......Page 208
   3.1.2 Machine Learning Approach......Page 209
   3.1.3 Hybrid Approach......Page 211
  4.1 Upper Ontology......Page 213
   4.1.1 Arabic WordNet Ontology......Page 215
   4.2.1 General Domains......Page 216
    Manual Approach......Page 217
    Statistical Approach......Page 219
    Linguistic Approach......Page 221
    Hybrid Approach......Page 222
    Uncategorized......Page 223
    Quran Ontology......Page 224
    Hadith Ontology......Page 225
 5 Conclusion......Page 226
Information Retrieval and Question Answering......Page 231
 Abstract......Page 232
 1 Introduction......Page 233
  2.1 A Novel Indexing Approach......Page 234
  2.2 The Significance Level of a Concept (SLC)......Page 236
  2.3 Semantic Distance Between Query and CS......Page 242
 3 System Architecture......Page 243
  4.1 Experimental Setup......Page 245
   4.2.1 The Conceptualization Levels......Page 247
   4.2.3 The Ranking Accuracy......Page 248
 5 Conclusion and Future Work......Page 249
 Appendix: The Implementation Algorithms......Page 250
 References......Page 253
An Empirical Study of Documents Information Retrieval Using DWT......Page 254
 1 Introduction......Page 255
  2.1 Term Signal......Page 256
  2.3 Document Segmentation......Page 257
  2.4 Wavelet Transform Algorithm......Page 258
  3.1 Problems and Design Issues......Page 261
  3.3 Document Segmentation......Page 262
  3.4 Term Weighting......Page 263
 4 Experiments and Results......Page 264
 References......Page 266
 Abstract......Page 268
 1 Introduction......Page 269
   2.1.1 Question Classification......Page 270
  2.2 Answer Type Determination......Page 273
  2.4 Query Expansion......Page 274
  2.5 Document Processing......Page 275
   2.5.1 Passage Retrieval......Page 277
   2.6.1 Named Entity Recognition......Page 278
 3 Developments in Hindi Question Answering System......Page 279
  3.1 Developments in Tasks of Question Answering Systems......Page 281
 4 Introduction to Hindi Language and Its Challenges for QASs......Page 282
 5 Tools and Resources for Hindi Question Answering......Page 283
 6 Future Scopes......Page 287
 References......Page 290
Text Classification......Page 296
 1 Introduction......Page 297
 2 Problem Definition......Page 299
  2.2 Problem Formalization......Page 301
  3.1 Data Selection and Preparation......Page 304
  3.2 Text Preprocessing......Page 306
  3.3 Document Indexing and Term Weighting Methods......Page 308
  3.4 Feature Reduction......Page 310
 4 Classification Algorithms......Page 314
 5 Arabic Text Classification......Page 317
 6 Directions for Further Research......Page 321
 7 Conclusion......Page 323
 References......Page 324
 1 Introduction......Page 327
  2.1 Authorship Attribution and NLP......Page 328
  2.2 Authorship Attribution in Arabic......Page 330
  3.2 Selection of Texts......Page 331
 4 Methods......Page 333
  4.1 JGAAP......Page 334
  4.2 Canonicizers......Page 335
  4.3 Event Drivers......Page 336
  4.4 Analysis Methods......Page 337
  5.1 Character n-grams......Page 338
  5.2 Word n-grams......Page 341
  5.4 Rare Words......Page 342
  5.5 Most Common Words......Page 343
 6 Analysis of Errors......Page 344
 7 Future Work and Conclusions......Page 348
 References......Page 349
 Abstract......Page 352
  2.2 Stop Words Removal......Page 353
  2.3 Word Stemming......Page 354
  2.5 Reuters 21,758 Test Collection for Text Categorization......Page 355
  2.6 Term Weighting Techniques......Page 356
  3.2 The Resulting Database Model......Page 357
  3.3 Weighting Techniques......Page 358
  3.4 Improvements for Weighting......Page 359
 4 Neural Network Based Classifier......Page 361
  4.1 Dimensionality Reduction for Text Categorization......Page 362
  4.2 The Proposed Neural Network Based Text Classifier......Page 365
  4.3 Experimental Details......Page 366
 6 Conclusions......Page 368
 References......Page 369
Text Mining......Page 371
 Abstract......Page 372
 1 Introduction......Page 373
  2.2 Information Extraction......Page 375
  2.4 Text Mining Methods and Techniques......Page 376
 3 Related Work......Page 378
  4.1 Text Mining Processing Framework......Page 381
  4.2 Data Collection and Pre-processing......Page 382
 5 Experimental Results......Page 383
 6 Conclusion......Page 392
 References......Page 393
 Abstract......Page 397
 1 Introduction......Page 398
 2 Literature Review......Page 399
 3 Research Methodology......Page 402
 4 Results......Page 403
 5 Conclusion......Page 411
 References......Page 412
 Abstract......Page 414
 1 Introduction......Page 415
  2.1 Text Mining Versus Data Mining......Page 416
 3 Text Mining of the Holy Quran......Page 417
 4 Sentiment Analysis of Arabic Text (Opinion Mining)......Page 419
 5 Text Mining of Arabic Web Documents......Page 421
 References......Page 426
Text Summarization......Page 429
TALAA-ATSF: A Global Operation-Based Arabic Text Summarization Framework......Page 430
 2 Related Work......Page 431
  2.2 Text Summarization Frameworks......Page 432
  2.3 Arabic Text Summarization......Page 433
 3 Challenges and Difficulties in Arabic Text Summarization......Page 435
  4.1 Definitions......Page 436
  5.1 General Idea......Page 440
  5.2 Why A Multi-layer Graph?......Page 441
  5.3 The Process of Summary Generation......Page 442
  6.1 Evaluation of the Algorithm for Affectation of Operations to Document Partitions......Page 446
  6.2 Evaluation of the Summarization Framework......Page 448
 7 Discussion......Page 449
 8 Conclusion......Page 450
 References......Page 451
 1 Introduction......Page 455
  2.1 Abstractive Summary......Page 456
  2.2 Extractive Summary......Page 457
  4.1 Topic-Word Summary......Page 458
  4.2 Centroid Summary......Page 460
  4.3 LexPageRank Summary......Page 461
  5.1 Centroid Summarizer......Page 463
  5.2 Topic-Word Summarizer......Page 464
  5.3 LexPageRank Summarizer......Page 465
  5.4 The Proposed Summarizer......Page 466
  7.1 Python 2.7.6......Page 467
  7.4 Rouge......Page 468
 8 Summarizer Performance......Page 469
 9 Conclusion......Page 470
 References......Page 471
Character and Speech Recognition......Page 473
 1 Introduction......Page 474
 3 General Principles on the Recognition of the On-line Writing......Page 476
 4 Time Delay Neural Networks TDNN......Page 478
 5 Preprocess......Page 482
 6 Experiments and Results......Page 485
 7 Discussion......Page 486
  7.1 Generalization Phase......Page 490
 8 Conclusion......Page 492
 References......Page 493
A Call Center Agent Productivity Modeling Using Discriminative Approaches......Page 494
 2 Related Work and Proposed Framework......Page 495
  2.1 Productivity Measurement Definition......Page 496
  2.2 Speech Recognition and Speaker Diarization......Page 497
  2.4 Sentiment Analysis and Binary Classification......Page 498
  3.1 Naïve Bayes Classifier (NB)......Page 499
  3.2 Logistic Regression (LR)......Page 501
  3.3 Linear Support Vector Machine (LSVM)......Page 505
 4 The Experiment......Page 509
 5 Results......Page 510
 6 Conclusion......Page 511
 References......Page 512
Morphological, Syntactic, and Semantic Processing......Page 514
 1 Introduction......Page 515
 2 Related Work......Page 516
 4 Development of the System Resources......Page 518
  4.1 Dictionary......Page 519
  4.2 The Linguistic Rules......Page 520
 5 Evaluation and Benchmarking......Page 532
 6 Conclusion......Page 533
 References......Page 534
 1 Introduction......Page 536
  2.1 Transition-Based Dependency Parsing......Page 540
  2.2 Graph-Based Dependency Parsing......Page 542
 3 Previous Work on Context Integration......Page 545
 4 RGB-2: A Context Integrating Dependency Parser......Page 546
  4.1 Context Representation......Page 547
  4.2 Feature Representation......Page 549
 5 System Description......Page 551
 6 Evaluation......Page 552
  6.1 Datasets......Page 553
  6.2 Experiments and Results......Page 554
 7 Conclusion and Future Work......Page 558
 References......Page 559
Fast, Accurate, Multilingual Semantic Relatedness Measurement Using Wikipedia Links......Page 561
 1 Introduction......Page 562
 2 Related Work......Page 563
 3 Similarity Assessment and Neighborhood Retrieval......Page 564
  4.1 Experimental Setting......Page 566
  4.2 Overall Relevance Assessment......Page 567
  4.3 Item by Item Relevance Assessment......Page 568
  5.1 Overall List Quality......Page 569
  5.2 Information Gain Analysis......Page 571
 References......Page 573
 1 Introduction......Page 575
 2 Related Works......Page 577
  3.1 Independent Language NLP Architectures......Page 578
  3.2 Arabic NLP Architectures......Page 582
  4.1 Comparison of Frameworks, Platforms and Toolkits......Page 586
  4.2 Benchmarking NLP Architectures......Page 587
 5 Discussion......Page 593
 6 Conclusion......Page 597
 References......Page 598
Building and Evaluating Linguistic Resources......Page 601
 1 Introduction......Page 602
 2 The Scope and Aims of Corpora......Page 604
 3 Stages in Corpus Building......Page 606
  3.1 Corpus Design and Compilation......Page 607
  3.2 Corpus Processing and Analysis......Page 608
 4 Overview of Progress in Arabic Corpus Linguistics......Page 609
  4.1 Quranic Corpora......Page 610
  4.2 Classical Arabic Corpora......Page 612
  4.3 Modern Standard Arabic Corpora......Page 613
  4.4 General Corpora......Page 615
  4.5 Multilingual Corpora Including Arabic Language......Page 616
  4.6 Dialectal Corpora......Page 617
  4.7 Tools and Analysers......Page 618
  4.8 Situation, Challenge and Recommendation......Page 619
 5 Conclusion......Page 620
 References......Page 621
An Evaluation of the Morphological Analysis of Egyptian Arabic TreeBank......Page 626
 1  Introduction......Page 627
 2 Related Studies......Page 628
 3  Social Media Usage in Arab World......Page 629
  4.1  Annotating Arabizi......Page 631
  5.1  Egyptian Arabic Morphological Analysis......Page 632
  5.2  Columbia Arabic Language Dialectal Morphological Analyzer CALIMA......Page 633
  5.3 Annotating ARZ ATB Corpus Using CALIMA......Page 634
  6.1  The Gold Standard......Page 635
  6.2 The Test Set......Page 637
  6.3 The Confusion Matrix......Page 639
  7.1 The Quantitative Analysis......Page 640
  7.2 Qualitative Analysis......Page 643
 References......Page 646
 1 Introduction......Page 648
 2 Related Works......Page 649
 3 Domain-Specific Comparable Corpora Building......Page 650
 4 Combined Anchor-Point-Based Method for Comparable Sentences Alignment......Page 651
 5 Compositional-Based Approach for Parallel Fragment Generation......Page 653
  6.1 Domain-Specific Comparable Corpora Evaluation......Page 655
  6.2 Comparable Sentences Evaluation......Page 656
  6.3 Parallel Phrases Evaluation......Page 658
 7 Conclusion......Page 662
 References......Page 663
 1 Introduction......Page 666
  2.2 Metadiscourse......Page 667
  2.4 Hedges......Page 668
  2.5 Meyer’s Taxonomy......Page 670
  2.6 Intensifiers......Page 671
  2.8 Bundles......Page 672
  2.10 Lexical and Functional Words......Page 673
  2.11 Corpus Linguistics Definition and Potential......Page 674
  3.2 Contrasting and Analysing the Two Corpora......Page 675
  3.4 Control Corpus Compilation......Page 676
  4.1 The Most Frequent Single Words......Page 678
  4.2 Lexical Density......Page 680
  4.3 Hyland’s Taxonomy......Page 681
  4.4 Salagar-Meyer’s Taxonomy......Page 682
  4.5 Syntactic and Semantic Tagging......Page 684
  4.6 Modals with Deontic and Epistemic Meanings......Page 686
  4.8 Intensifiers......Page 688
  4.9 State of Inexactitude......Page 690
  4.10 Collocational Frame (It Is … that)......Page 692
 5 Conclusion......Page 695
 References......Page 697
E-learning......Page 700
Intelligent Text Processing to Help Readers with Autism......Page 701
 1 Introduction......Page 702
 2 Related Work......Page 703
 3 Overview of the Project and the Language Technology......Page 704
  3.1 User Requirements......Page 705
  3.2 Architecture of the System......Page 706
  3.3 Processing Structural Complexity......Page 707
  3.4 Processing Ambiguity in Meaning......Page 712
  3.5 Generation of Personalised Documents......Page 715
  4.1 Reading Comprehension Testing......Page 717
  4.2 Evaluation of Text Conversion Using OpenBook: Readability Assessment......Page 719
  4.3 User Feedback......Page 721
 5 Discussion and Conclusions......Page 723
 References......Page 725
 1 Augmented Reality Technology......Page 729
  1.2 Augmented Reality (AR) Versus Virtual Reality (VR)......Page 730
  2.1 AR and Education......Page 731
  2.2 AR Web Application and Discovery-Based Learning......Page 733
  2.3 AR Medical Applications......Page 734
  2.5 Annotation and Visualization Manufacturing, Maintenance, and Repair......Page 735
  2.6 Engineering Design......Page 736
  2.9 Alive Journals......Page 737
  2.11 Smartphones Games and Application......Page 738
  2.12 Face Recognition and Personal Information Systems Using AR......Page 739
  2.13 Read Any Language......Page 740
  3.1 Display and Tracking System......Page 741
  3.2 Marker......Page 742
  3.3 Mobile Computing Power......Page 743
  4.2 Software Requirement......Page 744
  4.3 Hardware Requirements......Page 745
 References......Page 746
 1 Introduction......Page 748
  2.1 Automatic Query Expansion......Page 749
  2.4 Expansion Features Selection......Page 750
 4 Synthetic Dataset......Page 751
  5.1 Spectral Based Information Retrieval with QE Using KLD......Page 753
  5.2 Spectral Based Information Retrieval with QE Using WordNet......Page 757
 References......Page 762