Advances in computational intelligence. Part II. : 21st Mexican international conference on artificial intelligence, MICAI 2022, Monterrey, Mexico, October 24-29, 2022, proceedings / edited by Obdulia Pichardo Lagunas, Juan Martínez-Miranda, and Bella Martínez Seis.

Format
Book
Language
English
Published/​Created
  • Cham, Switzerland : Springer, [2022]
  • ©2022
Description
1 online resource (402 pages)

Details

Subject(s)
Editor
Series
Bibliographic references
Includes bibliographical references and index.
Source of description
Description based on print version record.
Contents
  • Intro
  • Preface
  • Conference Organization
  • Contents - Part II
  • Contents - Part I
  • Natural Language Processing
  • Urdu Named Entity Recognition with Attention Bi-LSTM-CRF Model
  • 1 Introduction
  • 2 Related Work
  • 2.1 Rule-Based Techniques
  • 2.2 Statistical Techniques
  • 2.3 Hybrid Techniques
  • 3 Our Proposal
  • 3.1 Embedding Layer
  • 3.2 Bi-LSTM Module
  • 3.3 Attention Layer
  • 3.4 Encoder Layer
  • 3.5 Conditional Random Field (CRF)
  • 3.6 Prediction Layer
  • 4 Experimental Setup
  • 4.1 Data
  • 4.2 Experiments
  • 4.3 Evaluation Metrics
  • 5 Results
  • 6 Conclusions
  • References
  • Impact Evaluation of Multimodal Information on Sentiment Analysis
  • 2 Dataset
  • 3 Method
  • 3.1 Text Processing and Feature Extraction
  • 3.2 Emoji Processing
  • 3.3 Text in Image Detection
  • 3.4 Image Sentiment Label
  • 3.5 Feature Fusion
  • 3.6 Classification Algorithm
  • 4 Experimental Results
  • 5 Conclusions and Future Work
  • .26em plus .1em minus .1emImproving Neural Machine Translation for Low Resource Languages Using Mixed Training: The Case of Ethiopian Languages
  • 1.1 High vs Low Resource Languages
  • 3 Dataset
  • 4 Methodology
  • 4.1 Data Pre-processing
  • 4.2 Model
  • 5 Experiments and Results
  • 5.1 Experiments
  • 5.2 Dataset Split
  • 5.3 Results
  • 6 Conclusions and Future Work
  • Machine Translation of Texts from Languages with Low Digital Resources: A Systematic Review
  • 2 Method
  • 2.1 Phase 1: Documentary Search
  • 2.2 Phase 2: Description of Selection Criteria
  • 2.3 Phase 3: Analysis and Categorization
  • 2.4 Phase 4: Discussion
  • 3 Conclusions
  • Comparison Between SVM and DistilBERT for Multi-label Text Classification of Scientific Papers Aligned with Sustainable Development Goals
  • 2 Related Work.
  • 2.1 Problem Transformation Method and Classification Algorithm
  • 2.2 Transfer Learning Model
  • 3 Methodology
  • 3.1 Framework for Multi-label Text Classification
  • 4 Model Experiments
  • 4.1 Dataset
  • 4.2 Data Preprocessing
  • 4.3 Models Building
  • 4.4 Model Evaluation
  • 5 Results and Discussion
  • 6 Conclusion and Future Work
  • A Hybrid Methodology Based on CRISP-DM and TDSP for the Execution of Preprocessing Tasks in Mexican Environmental Laws
  • 2 Methodology Conceptualization for the Preprocessing of Legislative Documents
  • 2.1 Methodology Description
  • 3 Methodology Implementation for the Preprocessing of Legislative Documents
  • 3.1 Judiciary of Mexico
  • 3.2 Text Preprocessing Use Case for Environmental Laws
  • 4 Validation of the Results
  • 4.1 Business Understanding Deliverables
  • 4.2 Data Understanding Deliverables
  • 4.3 Data Preparation Deliverables
  • 4.4 Experimental Process Deliverables
  • 4.5 Experimental Process Results
  • 5 Conclusions
  • News Intention Study and Automatic Estimation of Its Impact
  • 2 State of the Art
  • 2.1 Difference of This Proposal with Respect to the State of the Art Presented Above
  • 3 Solution Development
  • 3.1 Description of the Solution
  • 3.2 Scientific Novelty
  • 3.3 Evaluation
  • 4 Experiments and Results
  • Evaluation of a New Representation for Noise Reduction in Distant Supervision
  • 3.1 Dataset
  • 3.2 Baseline of Representation Learning Methods
  • 3.3 Unsupervised Representation Learning Methods
  • 3.4 Anomaly Detection Methods
  • 3.5 Experimental Design
  • 4 Experiments and Evaluation
  • Automatic Identification of Suicidal Ideation in Texts Using Cascade Classifiers
  • 1 Introduction.
  • 2 Related Works
  • 3 Proposed Method
  • 3.1 Datasets
  • 3.2 Weighing of Terms
  • 3.3 Cascade Classification
  • 4 Experimentation and Results
  • Web Crawler and Classifier for News Articles
  • 2 Web Crawling and Corpus Creation
  • 3 Classifier
  • 4 Web Application
  • Sentiment Analysis in the Rest-Mex Challenge
  • 2 Corpus and Task Description
  • 3.1 Lexicon-Based Approach
  • 3.2 Machine Learning Approach
  • 4.1 Experiments with the Lexicon-Based Approach
  • 4.2 Experiments with the Machine Learning Approach
  • A Bibliometric Review of Methods and Algorithms for Generating Corpora for Learning Vector Word Embeddings
  • 2 Methodology
  • 2.1 Methods
  • 2.2 Materials
  • 3 Results
  • 3.1 Publications and Citations by Year
  • 3.2 Top Conferences and Journals
  • 3.3 Top Institutions and Funding Organizations
  • 3.4 Top Influential Publications
  • 3.5 Top Countries by Publications
  • 4 Discussion
  • 5 Conclusion
  • Evaluating the Impact of OCR Quality on Short Texts Classification Task
  • 3.1 Selection of Classification Categories
  • 3.2 Obtaining OCRed Text
  • 3.3 The Human Typed Text and Its Approximations
  • 3.4 Train/Test Split
  • 4 Classification Approaches
  • 4.1 Fuzzy Substring Search
  • 4.2 CNN
  • 4.3 BERT
  • 4.4 RoBERTa
  • 5.1 Evaluation Metrics
  • 5.2 Experiments
  • Techniques for Generating Language Learning Resources: A System for Generating Exercises for the Differentiation of Literal and Metaphorical Context
  • 2 Characteristics of Automatic Generation Systems for Learning.
  • 2.1 NLP Tasks at Automatic Generation Systems for Language Learning
  • 2.2 Creation Sources
  • 2.3 Technical and Methodological Aspects of Automatic Generation Systems for Learning
  • 2.4 Text Preprocessing
  • 3 An Automatic Generation System for Learning the English Language: Literal Language and Metaphorical Language
  • 3.1 Corpus Selection
  • 3.2 System Development: General Preprocessing
  • 3.3 System Development: Exercises
  • 4 Results
  • 4.1 The Exercises Generated and Their Validation
  • Exploratory Data Analysis for the Automatic Detection of Question Paraphrasing in Collaborative Environments
  • 2 Theorical Framework
  • 3 State of the Art
  • 3.1 Scientific Novelty and Difference with the State of the Art
  • 4 Solution Development
  • 4.1 Initial EDA
  • 4.2 Preprocessing
  • 4.3 Feature Extraction
  • 4.4 Analysis of Preprocessed Data
  • 4.5 Classifier Models
  • 4.6 Model Comparison
  • Best Paper Award
  • Diachronic Neural Network Predictor of Word Animacy
  • 2 Data and Method
  • 4 Case Study
  • Sequential Models for Sentiment Analysis: A Comparative Study
  • 2 Literature Review
  • 3 Experimental Setup
  • 4 Results and Discussion
  • Intelligent Applications and Robotics
  • Analysis of Procedural Generated Textures for Video Games Using a CycleGAN
  • 2.1 Procedural Content Generation
  • 2.2 Generative Adversarial Networks
  • 2.3 CycleGAN
  • 3 Proposed Approach
  • 4.1 CycleGAN Training Details
  • 4.2 CycleGAN Results
  • 4.3 Emotional Data Collection
  • 4.4 Emotional Results
  • References.
  • Vibration Analysis of an Industrial Motor with Autoencoder for Predictive Maintenance
  • 1.1 A Subsection Sample
  • 2 Background
  • 3.1 Materials and Test Bench Construction
  • 6 Conclusions and Future Work
  • Modeling and Simulation of Swarm of Foraging Robots for Collecting Resources Using RAOI Behavior Policies
  • 2 Foraging Behavior
  • 3 Swarm Features
  • 3.1 Robot Mathematical Model
  • 3.2 Limitations of Sensory Capacity of Robots
  • 3.3 Local Behavior Rules
  • 4 Simulation Environment
  • 6 Conclusion
  • Data-Driven Adaptive Force Control for a Novel Soft-Robot Based on Ultrasonic Atomization
  • 2 Actuation and Soft-Robot Design
  • 2.1 Ultrasonic Atomization
  • 2.2 Rapid Actuator Design
  • 2.3 Mini Soft-Robot Design
  • 3 Intelligent Control
  • 3.1 Neuro-Fuzzy Architecture
  • 3.2 Adaptation Algorithm
  • Data-driven-modelling and Control for a Class of Discrete-Time Robotic System Using an Adaptive Tuning for Pseudo Jacobian Matrix Algorithm
  • 2 Data-driven Model for Discrete-Time System
  • 2.1 Equivalent Model Stability Analysis
  • 2.2 Neuro-fuzzy Network and Adaptive Step Parameter
  • 3 Control Law
  • Retrieval-based Statistical Chatbot in a Scientometric Domain
  • 2.1 Natural Language Processing
  • 2.2 Intent Classificaiton and Entity Extraction
  • 2.3 Data Labeling
  • 2.4 Model Training
  • 2.5 Scientometric Indicator Identification
  • 2.6 Natural Language Transformation into Cypher Query
  • 2.7 Chatbot Deployment
  • 3 Results and Discussion
  • 3.1 Goal Completion Rate
  • 3.2 Survey Evaluation
  • 4 Conclusion
  • Red Light/Green Light: A Lightweight Algorithm for, Possibly, Fraudulent Online Behavior Change Detection.
ISBN
3-031-19496-9
Statement on language in description
Princeton University Library aims to describe library materials in a manner that is respectful to the individuals and communities who create, use, and are represented in the collections we manage. Read more...
Other views
Staff view

Supplementary Information