Mónica Bécue-Bertaut: Textual and Contextual Data Analysis, Gebunden

Klappentext

Multidimensional statistical analysis of textual data is a powerful technique that enables researchers to uncover deeper insights into the context and meaning of documents. This book addresses the challenge of jointly analyzing textual and contextual data, presenting rigorous theoretical foundations alongside practical methodologies. By incorporating metadata and contextual information, readers can extract richer, more nuanced information from textual corpora, making this book an essential resource for statisticians, data scientists, and linguistics experts.

The book explores a wide range of textual data, from open-ended survey responses and political speeches to legal texts, literary works, and technical reports. It also examines the diverse contextual variables that shape these texts, such as sociodemographic characteristics, chronology, political affiliations, and external influences. Through real-world examples, readers will learn how to apply exploratory multivariate statistical methods to compare, characterize, and reveal the underlying structure of textual data. Each chapter builds on the previous one, offering a systematic approach to encoding, analyzing, and visualizing textual and contextual data. Topics include machine learning methods like latent semantic analysis and correspondence analysis, clustering techniques, restricted clustering defined by contextual data, and advanced visualization tools. The book also introduces methodologies for analyzing multilingual corpora and isolated texts, emphasizing the importance of discourse strategies and thematic contrasts.

This book is not only a guide to advanced statistical methods but also a practical toolkit for researchers working with diverse corpora. Whether analyzing legal databases, sensory evaluations, or political speeches, readers will find robust techniques to uncover patterns, relationships, and strategies within their data. By combining textual and contextual analysis, this book empowers readers to make meaningful comparisons and draw actionable conclusions.

KEY FEATURES:

Comprehensive coverage of methods for jointly analyzing textual and contextual data.

Practical applications to diverse corpora, including legal texts, political speeches, and sensory evaluations.

Systematic comparison of machine learning methods like latent semantic analysis and correspondence analysis.

Advanced visualization techniques, including interactive, 3D, and animated graphics.

Methodologies for analyzing multilingual corpora and isolated texts, with a focus on discourse strategies.