東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Ideology, Social Media and Fake News: New Machine Learning Methods for Political Science.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Ideology, Social Media and Fake News: New Machine Learning Methods for Political Science./
作者:	Godel, William.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2022,
面頁冊數:	174 p.
附註:	Source: Dissertations Abstracts International, Volume: 84-01, Section: A.
Contained By:	Dissertations Abstracts International84-01A.
標題:	Political science. -
ISBN:	9798837544682

Ideology, Social Media and Fake News: New Machine Learning Methods for Political Science.
Godel, William.

Ideology, Social Media and Fake News: New Machine Learning Methods for Political Science. - Ann Arbor : ProQuest Dissertations & Theses, 2022 - 174 p.

Source: Dissertations Abstracts International, Volume: 84-01, Section: A.

Thesis (Ph.D.)--New York University, 2022.

This item must not be sold to any third party vendors.

New machine learning tools have created many opportunities to address outstanding questions in the field. This dissertation introduces novel machine learning methods and applications to address both substantive and methodological challenges in political science. Chapter one introduces a novel unsupervised algorithm that measures politician ideology on a circle, which orders politicians in a more coherent ordering than current methods. This method is simple to implement, fast, and robust to specification, while producing estimates that are extremely similar to alternative approaches. Chapter two introduces a new two stage machine learning pipeline to estimate politicians' expressed ideology dynamically, based on their tweets. Using this method, the chapter then shows that some politicians do moderate in their expressed ideology over the course of a U.S. two stage election campaign. Finally, Chapter three attempts to identify if "True" new can be identified from low credibility sources utilizing crowds of lay respondents. Using both simple rules, such as the mode of the crowd, and more complex machine learning based methods, it shows that while crowd sourcing can perform better than guessing, it can not reach the level of performance equal to that of a professional fact checker.

ISBN: 9798837544682Subjects--Topical Terms:

528916
Political science.
Subjects--Index Terms:

Ideology

Ideology, Social Media and Fake News: New Machine Learning Methods for Political Science.
LDR:02372nmm a2200361 4500 001 2349694
005 20221003074932.5
008 241004s2022 eng d
020 $a 9798837544682
035 $a (MiAaPQ)AAI28968908
035 $a AAI28968908
040 $a MiAaPQ $c MiAaPQ
100 1 $a Godel, William. $3 3689110
245 1 0 $a Ideology, Social Media and Fake News: New Machine Learning Methods for Political Science.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2022
300 $a 174 p.
500 $a Source: Dissertations Abstracts International, Volume: 84-01, Section: A.
500 $a Advisor: Spirling, Arthur;Dragu, Tiberiu.
502 $a Thesis (Ph.D.)--New York University, 2022.
506 $a This item must not be sold to any third party vendors.
520 $a New machine learning tools have created many opportunities to address outstanding questions in the field. This dissertation introduces novel machine learning methods and applications to address both substantive and methodological challenges in political science. Chapter one introduces a novel unsupervised algorithm that measures politician ideology on a circle, which orders politicians in a more coherent ordering than current methods. This method is simple to implement, fast, and robust to specification, while producing estimates that are extremely similar to alternative approaches. Chapter two introduces a new two stage machine learning pipeline to estimate politicians' expressed ideology dynamically, based on their tweets. Using this method, the chapter then shows that some politicians do moderate in their expressed ideology over the course of a U.S. two stage election campaign. Finally, Chapter three attempts to identify if "True" new can be identified from low credibility sources utilizing crowds of lay respondents. Using both simple rules, such as the mode of the crowd, and more complex machine learning based methods, it shows that while crowd sourcing can perform better than guessing, it can not reach the level of performance equal to that of a professional fact checker.
590 $a School code: 0146.
650 4 $a Political science. $3 528916
650 4 $a Journalism. $3 576107
650 4 $a Rhetoric. $3 516647
650 4 $a Web studies. $3 2122754
653 $a Ideology
653 $a Social media
653 $a Fake news
653 $a Machine learning
690 $a 0615
690 $a 0646
690 $a 0391
690 $a 0681
710 2 0 $a New York University. $b Politics. $3 1535717
773 0 $t Dissertations Abstracts International $g 84-01A.
790 $a 0146
791 $a Ph.D.
792 $a 2022
793 $a English