| ISBN： | 9780691207544 : |
|---|
| ISBN： | 9780691207551 |
|---|
| 编目源： | UKMGB UKMGB OCLCO GZN OCLCF USD YDX PIT OCLCO GUA MBB SFB QGQ NWQ EAU |
|---|
| 个人名称： | Grimmer, Justin, |
|---|
| 题名： | Text as data : a new framework for machine learning and the social sciences / Edited by Justin Grimmer, Margaret E. Roberts, Brandon M. Stewart. |
|---|
| 出版发行项： | Princeton, New Jersey : Princeton University Press, [2022] |
|---|
| 载体形态： | xix, 336 pages : illustrations ; 26 cm |
|---|
| 书目附注： | Includes bibliographical references (pages [307]-329) and index. |
|---|
| 格式化内容附注： | Part I. Preliminaries. Introduction ; Social science research and text analysis -- Part II. Selection and representation. Principles of selection and representation ; Selecting documents ; Bag of words ; The multinominal language model ; The vector space model and similarity metrics ; Distributed representations of words ; Representations from language sequences -- Part III. Discovery. Principles of discovery ; Discriminating words ; Clustering ; Topic models ; Low-dimensional document embeddings -- Part IV. Measurement. Principles of measurement ; Word counting ; An overview of supervised classification ; Coding a training set ; Classifying documents with supervised learning ; Checking performance -- Repurposing discovery methods -- Part V. Inference. Principles of inference ; Prediction ; Casual inference ; Text as outcome ; Text as treatment ; Text as confounder -- Part VI. Conclusion. |
|---|