Text as Data: A New Framework for Machine Learning and the Social Sciences

Researchers can gain new insights into the social world by analyzing large collections of text using computational text analysis. Text as Data provides a guide to combining new data sources, machine learning tools, and social science research design to develop and evaluate insights.

Format: Paperback / softback
Length: 360 pages
Publication date: 29 March 2022
Publisher: Princeton University Press

From social media posts and text messages to digital government documents and archives, researchers are inundated with a flood of text that reflects the social world. This textual data provides unprecedented insights into fundamental questions in the social sciences, humanities, and industry. Meanwhile, new machine learning tools are rapidly transforming the way science and business are conducted.

Text as Data: Unlocking New Insights with Computational Text Analysis provides a comprehensive guide for leveraging computational text analysis to explore the social world. The book is organized around the core tasks involved in research projects using text, including representation, discovery, measurement, prediction, and causal inference. The authors offer a sequential, iterative, and inductive approach to research design, presenting each research task with real-world applications, example methods, and a distinct style of task-focused research.

One of the key strengths of Text as Data is its ability to bridge the gap between computer science and social science. The book emphasizes the importance of combining new sources of data, such as social media and text messages, with traditional social science research methods, such as surveys and interviews. This approach allows researchers to tap into the vast amounts of textual data available and gain a deeper understanding of complex social phenomena.

Another notable feature of the book is its emphasis on practical applications. Each research task is presented with real-world examples from across the social sciences and industry, demonstrating how computational text analysis can be used to address pressing social problems and advance scientific knowledge. The authors provide detailed instructions and code examples to help readers implement these methods in their own research projects.

In conclusion, Text as Data is an essential resource for anyone interested in analyzing large collections of text in an era where data is abundant and computation is cheap. The book provides a comprehensive and practical guide to leveraging computational text analysis to gain new insights into the social world. Whether you are a social scientist, humanities researcher, or industry professional, this book will help you unlock the power of text as a data source and advance your research agenda.

Weight: 718g
Dimension: 177 x 253 x 25 (mm)
ISBN-13: 9780691207551