Language text data models (5 Ergebnisse)

Zustand: As New. Unread book in perfect condition.

Hardcover. Zustand: Near Fine. 1st Edition. Hardcover, vi + 280 pages, NOT ex-library. Minor handling wear only, book is fresh, clean and bright throughout with unmarked text, free of inscriptions and stamps, firmly bound. Issued without a dust jacket. -- Hardcover, vi + 280 pages, NOT ex-library. Minor handling wear only, book is fresh, clean and bright throughout with unmarked text, free of inscriptions and stamps, firmly bound. Issued without a dust jacket. -- A collection of research in quantitative linguistics and text mining. It brings together international scholars to demonstrate how mathematical models and statistical data can be used to uncover the underlying structures of human language. The book is dedicated to the memory of Ludek Hrebícek, a pioneer in the field, and reflects his influence on treating language as a dynamic, self-organising system. Quantitative Models: The contributors apply various statistical laws (such as Zipf's Law and the Menzerath-Altmann Law) to analyse word frequency, sentence length, and the distribution of linguistic units. Textual Dynamics: Several chapters explore "textual space," examining how information is distributed throughout a document and how different parts of a text relate to one another mathematically. Information Theory: The book uses concepts like entropy and information density to measure the complexity of different languages and genres. Applied Linguistics: Beyond pure theory, the text demonstrates how these models can be applied to authorship attribution (identifying who wrote a text), stylometry, and machine translation. The research is categorised into several specialised fields: - Lexicology and Grammar: Statistical analysis of vocabulary growth and the hierarchical structure of grammatical units. - Dialectology and Typology: Using data-driven methods to compare different languages and regional dialects. - History of Linguistics: Essays reflecting on the evolution of quantitative methods from the mid-20th century to the era of Big Data. - Digital Humanities: Demonstrations of how computational tools can process vast corpora of historical and contemporary texts to find patterns invisible to the naked eye. The book is highly relevant for researchers in natural language processing (NLP) and corpus linguistics. It provides a rigorous theoretical foundation for the algorithms that power modern search engines and AI language models. …

Taschenbuch. Zustand: Neu. Neuware - Unlock the Power of PyTorch 2.0 for Next-Level Natural Language Processing.

Taschenbuch. Zustand: Neu. Neuware - Reactive PublishingEconometrics of Text and Large Language Models with Python presents practical methods for applying econometric techniques to unstructured text data using modern language models.The book focuses on three core tasks: measuring latent concepts from text, generating predictions from textual sources, and conducting causal inference with unstructured data. It shows how to move from raw text to reliable quantitative results while maintaining the standards of econometric rigor.Readers will find clear explanations of the underlying statistical principles alongside working Python implementations. Topics include text representation, embedding-based measurement, predictive modeling with language models, and strategies for identifying causal effects when the primary data source is text. Emphasis is placed on validation, robustness checks, and the practical limits of these methods.This book is written for applied researchers, data scientists, and quantitative analysts who work with textual data and need reproducible, transparent approaches grounded in econometrics. It assumes familiarity with basic statistics and Python but does not require prior expertise in natural language processing or large language models.By combining established econometric frameworks with current language model tools, the book provides a structured path for extracting meaningful, verifiable insights from unstructured text.…

Hardcover. Zustand: Brand New. 280 pages. 9.50x6.50x0.75 inches. In Stock.