Statistical Models and Learning Methods for Complex Data
Description
This book on statistical models and learning methods for complex data comprises a selection of peer-reviewed post-conference papers presented at the 14th Scientific Meeting of the Classification and Data Analysis Group of the Italian Statistical Society (CLADAG 2023), held in Salerno, Italy, September 11-13, 2023. The contributions span a variety of topics, including different approaches to clustering and classification, multidimensional data analysis, panel data, social networks, time series, statistical inference, and mixture models. These methodologies are applied to a range of empirical domains such as economics, finance, hydrology, the social sciences, education, and sports.
Organized biennially by international scientific committees, the CLADAG meetings advance methodological research in multivariate statistics, with a strong focus on data analysis and classification. They facilitate the exchange of ideas in these fields and promote the dissemination of concepts, numerical methods, algorithms, and computational and applied results.
Chapter "Identification of misogynistic accounts on Twitter through Graph Convolutional Networks" is available open access under a Creative Commons Attribution 4.0 International License via link.springer.com.
More details
Other editions
Additional editions
Persons
Giuseppe Giordano is an Associate Professor of Statistics at the University of Salerno, Italy. He teaches in the area of statistics and its applications in the social sciences for bachelor's and master's degrees, and Ph.D. programs. His research interests are mainly in multidimensional data analysis, mortality data tables, social network analysis, preference data, and textual data analysis, emphasising the development and use of statistical methods in applied contexts.
Michele La Rocca is a Full Professor of Statistics at the University of Salerno in Italy. His research areas include empirical likelihood, nonlinear time series, neural networks and deep learning, and resampling techniques, which he has applied to biological and financial data. A crucial aspect of his research is building bridges between data analysis, statistics, and machine learning methods. He is an elected member of ISI and the Charting Committee of the International Symposia on Nonparametric Statistics (ISNPS), an IMS group. He has served as Associate Editor of Computational Statistics and Data Analysis and is now Associate Editor of Statistical Methods and Applications, the journal of the Italian Statistical Society.
Marcella Niglio is an Associate Professor of Statistics at the University of Salerno, Italy. She teaches elements of statistical analysis, statistical modelling, advanced statistical inference, and data visualisation and reporting for bachelor's and master's degrees, and Ph.D. programs. Her research interests are related to time series analysis, and regression models with imbalanced data, whereas the theoretical issues are mainly applied in economics, finance, and in educational contexts. She is a member of the Italian Statistical Society.
Marialuisa Restaino is an Associate Professor of Statistics at the University of Salerno, Italy. Her teaching activity focuses on introductory statistics, computational statistics, and statistical modelling for both bachelor's and master's degrees, and Ph.D. programs. Her research interests relate to survival analysis, multistate models, and regression models in the presence of imbalanced datasets. Her methodological works are joined with applications for evaluating students' performance and predicting business failure. She is a member of the Italian Statistical Society, and of the International Biometric Society - Italian Region.
Maurizio Vichi is a Full Professor of Statistics at the Sapienza University of Rome, Italy. He is the founding President of the Federation of European National Statistical Societies, and a former President of the Italian Statistical Society, of the International Federation of Classification Societies, and Chair of the European Statistical Advisory Committee of the EU. He is also Coordinating Editor of Advances in Data Analysis and Classification and member of the European Statistical Governance Advisory Body of the EU. He teaches Multivariate Statistics, and Advances in Data Analysis and Statistical Modelling. He is the author of more than 150 papers, mainly on new multivariate statistical methods for data analysis, including big data and statistical tools for decision-making.
Content
- Exploring latent evolving ability in test equating and its effects on final rankings.- Hidden Markov and related discrete latent variable models An application to compositional data.- An application of Natural Language Processing Analysis on TripAdvisor Reviews.- Modelling football players field position via mixture of Gaussians with flexible weights.- Estimation Issues in Multivariate Panel Data.- Testing linearity in the single functional index model for dependent data.- A multi-step approach for streamflow classification.- Identification of misogynistic accounts on Twitter through Graph Convolutional Networks.- Topic modeling of publication activity in Hungary and Poland in the fields of economics, finance, and business.- Circular kernel classification with errors-in-variables.- Classification Trees Applied to Time Lagged Data to Improve Quality in Official Statistics.- Trimmed factorial k-means a clustering application to a cookies dataset_Farné and Camillo.- Visualization of Proximity and Role-based Embeddings in a Regional Labour Flow Network.- Bridging the Gap Investigating Correlation Clustering and Manifold Learning Connections.- Improving Performance in Neural Networks by Dendrite-Activated Connection.- Regression models with compositional regressors in case of structural zeros.- Multi-Dimensional Robinson Dissimilarities.- Composite selection criteria for the number of components of a finite mixture for ordinal data.- Clustering of Italian higher education institutions based on a destination-specific approach.- Analyzing Italian crime data using matrix-variate hidden Markov models.
