Data Science, Classification, and Artificial Intelligence for Modeling Decision Making
Produktdetaljer
Biographical note
Javier Trejos is a Full Professor and Researcher at the School of Mathematics and the Center for Research in Pure and Applied Mathematics (CIMPA), University of Costa Rica. His research focuses on the relations between data analysis and combinatorial optimization. He was the chief editor of the Journal of Mathematics: Theory and Applications and is the former president of the Central American and Caribbean Society for Classification and Data Analysis (SoCCCAD). In 1996 he was awarded the Simon Régnier Prize of the Francophone Classification Society.
Theodore Chadjipandelis is Professor of Applied Statistics and the Director of the Laboratory of Applied Political Research, Aristotle University, Thessaloniki, Greece. His research interests are in the field of applied statistics and mainly focus on issues of experiment design, statistical research training, public opinion, political and electoral behavior, electoral geography, election systems as well as urban and regional programming and development. He coordinated the Greek section of the program C.C.S. (Comparative Candidates Survey) – a co-operation between 30 research teams – and of C.S.E.S. (Comparative Study of Electoral Systems). Currently he coordinates the Greek section of the program MeDem (Measuring Electoral Democracy) - a co-operation between 30 research teams - and of the Horizon project AI4GOV (Artificial Intelligence for Governance).
Aurea Grané is Full Professor of Statistics and Operations Research at Universidad Carlos III de Madrid, Spain. Her work involves several lines of research whose common link is the development of non-parametric techniques based on distances with application to data of a certain complexity. She has important contributions in the development of goodness-of-fit statistics for uniformity, exponentiality and normality tests, in statistical methods based on distances for data visualization, in predictive methods for functional data and in the development of tools for outlier detection in long financial series and mixed data sets.
Mario Villalobos is a Professor and Researcher at the University of Costa Rica, School of Mathematics, and the Center for Research in Pure and Applied Mathematics (CIMPA), of which he was its director until 2020, and a lecturer at the Costa Rica Institute of Technology. His research deals mainly with multi-objective optimization and its relationships with statistical and data analysis methods, the study of functions, teaching innovations in mathematics, and currently curve-fitting to see trends in epidemics. He was the recipient of the Chikio Hayashi Award from the International Federation of Classification Societies in 2006.
This book gathers selected and peer-reviewed contributions presented at the 18th Conference of the International Federation of Classification Societies (IFCS 2024), held in San José, Costa Rica, July 15–19, 2024. Covering a wide range of topics, it describes modern methods and real-world applications in data science, classification, and artificial intelligence related to modeling decision making.
Numerous novel techniques and innovative applications are investigated, such as anomaly detection in public procurement processes, multivariate functional data clustering, air pollution prediction, benchmark generation for probabilistic planning, recommendation systems based on symbolic data analysis, and methods for clustering mixed-type data. Advanced statistical concepts are explored, including Vapnik-Chervonenkis dimensionality, Riemannian statistics, hypothesis testing for interval-valued data, and mixed models. Furthermore, machine learning techniques are applied to predict soil bacterial and fungal communities, classify electoral behavior and political competition, and assess corrosion degradation in mining pipelines.
The diversity of topics discussed in this collection reflects the ongoing advancement and interdisciplinary nature of statistical and data science research, as well as its application across various fields and sectors. These studies contribute to the development of robust methodologies and efficient computational tools to address complex challenges in the era of big data.
The book is intended for researchers and practitioners seeking the latest developments and applications in the field of data science and classification.
Preface.- Acknowledgements.- G. Afriyie, D. Hughes, A. Nettel Aguirre, N. Li, C. H. Lee, L. M. Lix, and T. Sajobi: A Comparison of Multivariate Mixed Models and Generalized Estimation Equations Models for Discrimination in Multivariate Longitudinal Data.- C. Adela Anton and I. Smith: A Multivariate Functional Data Clustering Method Using Parsimonious Cluster Weighted Models.- J. P. Arroyo-Castro and S. W. Chou-Chen: Unsupervised Detection of Anomaly in Public Procurement Processes.- Z. Aouabed, M. Achraf Bouaoune, V. Therrien, M. Bakhtyari, M. Hijri, and V. Makarenkov: Predicting Soil Bacterial and Fungal Communities at Different Taxonomic Levels Using Machine Learning.- V. Bouranta, G. Panagiotidou and T. Chadjipadelis: Candidates, Parties, Issues and the Political Marketing Strategies: A Comparative Analysis on Political Competition in Greece.- J. Cervantes, M. Monge, and D. Sabater: Predicting Air Pollution in Beijing, China Using Chemical, and Climate Variables.- J. Champagne Gareau, É. Beaudry, and V. Makarenkov: Towards Topologically Diverse Probabilistic Planning Benchmarks: Synthetic Domain Generation for Markov Decision Processes.- P. Chaparala and P. Nagabhushan: Symbolic Data Analysis Framework for Recommendation Systems: SDA-RecSys.- E. Costa, I. Papatsouma, and A. Markos: A Deterministic Information Bottleneck Method for Clustering Mixed-Type Data.- M. Farnia and N. Tahiri: A New Metric to Classify B Cell Lineage Tree.- T. Górecki, M.Krzyśko, and W. Wolyński: Applying Classification Methods for Multivariate Functional Data.- K. Moussa Sow and N. Ghazzali: Machine Learning-Based Classification and Prediction to Assess Corrosion Degradation in Mining Pipelines.- G. Nason, D. Salnikov, and M. Cortina-Borja: Modelling Clusters in Network Time Series with an Application to Presidential Elections in the USA.- M. A. Nunez and M. A. Schneider: On the Vapnik-Chervonenkis Dimension and Learnability of the Hurwicz Decision Criterion.- W. Pan and L. Billard: Distributional-based Partitioning with Copulas.- G. Panagiotidou and T. Chadjipadelis: Mapping Electoral Behavior and Political Competition: A Comparative Analytical Framework for Voter Typologies and Political Discourses.- O. Rodríguez Rojas: Riemannian Statistics for Any Type of Data.- A. Roy and F. Montes: Hypothesis Testing of Mean Interval for p-dimensional Interval-valued Data.- M. Solís and A. Hernández: UMAP Projections and the Survival of Empty Space: A Geometric Approach to High-Dimensional Data.- Q. Stier and M. C. Thrun: An Efficient Multicore CPU Implementation of the DatabionicSwarm.
This book gathers selected and peer-reviewed contributions presented at the 18th Conference of the International Federation of Classification Societies (IFCS 2024), held in San José, Costa Rica, July 15–19, 2024. Covering a wide range of topics, it describes modern methods and real-world applications in data science, classification, and artificial intelligence related to modeling decision making.
Numerous novel techniques and innovative applications are investigated, such as anomaly detection in public procurement processes, multivariate functional data clustering, air pollution prediction, benchmark generation for probabilistic planning, recommendation systems based on symbolic data analysis, and methods for clustering mixed-type data. Advanced statistical concepts are explored, including Vapnik-Chervonenkis dimensionality, Riemannian statistics, hypothesis testing for interval-valued data, and mixed models. Furthermore, machine learning techniques are applied to predict soil bacterial and fungal communities, classify electoral behavior and political competition, and assess corrosion degradation in mining pipelines.
The diversity of topics discussed in this collection reflects the ongoing advancement and interdisciplinary nature of statistical and data science research, as well as its application across various fields and sectors. These studies contribute to the development of robust methodologies and efficient computational tools to address complex challenges in the era of big data.
The book is intended for researchers and practitioners seeking the latest developments and applications in the field of data science and classification.Produktdetaljer
Biographical note
Javier Trejos is a Full Professor and Researcher at the School of Mathematics and the Center for Research in Pure and Applied Mathematics (CIMPA), University of Costa Rica. His research focuses on the relations between data analysis and combinatorial optimization. He was the chief editor of the Journal of Mathematics: Theory and Applications and is the former president of the Central American and Caribbean Society for Classification and Data Analysis (SoCCCAD). In 1996 he was awarded the Simon Régnier Prize of the Francophone Classification Society.
Theodore Chadjipandelis is Professor of Applied Statistics and the Director of the Laboratory of Applied Political Research, Aristotle University, Thessaloniki, Greece. His research interests are in the field of applied statistics and mainly focus on issues of experiment design, statistical research training, public opinion, political and electoral behavior, electoral geography, election systems as well as urban and regional programming and development. He coordinated the Greek section of the program C.C.S. (Comparative Candidates Survey) – a co-operation between 30 research teams – and of C.S.E.S. (Comparative Study of Electoral Systems). Currently he coordinates the Greek section of the program MeDem (Measuring Electoral Democracy) - a co-operation between 30 research teams - and of the Horizon project AI4GOV (Artificial Intelligence for Governance).
Aurea Grané is Full Professor of Statistics and Operations Research at Universidad Carlos III de Madrid, Spain. Her work involves several lines of research whose common link is the development of non-parametric techniques based on distances with application to data of a certain complexity. She has important contributions in the development of goodness-of-fit statistics for uniformity, exponentiality and normality tests, in statistical methods based on distances for data visualization, in predictive methods for functional data and in the development of tools for outlier detection in long financial series and mixed data sets.
Mario Villalobos is a Professor and Researcher at the University of Costa Rica, School of Mathematics, and the Center for Research in Pure and Applied Mathematics (CIMPA), of which he was its director until 2020, and a lecturer at the Costa Rica Institute of Technology. His research deals mainly with multi-objective optimization and its relationships with statistical and data analysis methods, the study of functions, teaching innovations in mathematics, and currently curve-fitting to see trends in epidemics. He was the recipient of the Chikio Hayashi Award from the International Federation of Classification Societies in 2006.