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

Improving Text Representations With ...

Lian, Ruixue.

FindBook

Google Book

Amazon

博客來

Improving Text Representations With Language Models: Contrastive Learning, Personalization, and Data Augmentation.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Improving Text Representations With Language Models: Contrastive Learning, Personalization, and Data Augmentation./
作者:	Lian, Ruixue.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2023,
面頁冊數:	117 p.
附註:	Source: Dissertations Abstracts International, Volume: 85-03, Section: A.
Contained By:	Dissertations Abstracts International85-03A.
標題:	Electrical engineering. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30637304
ISBN:	9798380142151

Improving Text Representations With Language Models: Contrastive Learning, Personalization, and Data Augmentation.
Lian, Ruixue.

Improving Text Representations With Language Models: Contrastive Learning, Personalization, and Data Augmentation. - Ann Arbor : ProQuest Dissertations & Theses, 2023 - 117 p.

Source: Dissertations Abstracts International, Volume: 85-03, Section: A.

Thesis (Ph.D.)--The University of Wisconsin - Madison, 2023.

Text representation learning is a fundamental task in NLP. It aims to capture the semantic, syntactic, and contextual information present in textual data and transform it into a numerical representation that machine learning models can effectively utilize. Effective text representations are crucial for various downstream tasks such as text classification, information retrieval, machine translation, and question answering, etc.This thesis presents various methodologies to improve sentence-level text representations with language models. Given the fact that subcategories under the same branches are similar and closer to each other, we propose variations of label-aware supervised contrastive loss (LA-SCL) to incorporate and learn the label hierarchy information. Two studies were conducted to improve text representations with personalized features, which can be tailored to specific users by incorporating individual information such as user demographics, or contextual factors. A novel model architecture called personalized transformer memory (PersonalTM) was proposed to effectively involve personalized features given a transformer-based encoder-decoder model on information retrieval tasks. An incremental learning algorithm was proposed to continuously keep the model up-to-date by incorporating streaming personalized data. A study was conducted to perform data augmentation by utilizing GPT-2 to generate text data, specifically focusing on producing utterances that are similar to original data in both semantics and sentiment.Several practical tasks in Social Media Analysis are also conducted in this thesis: several language models were employed to assess the favorability and hesitancy towards COVID-19 on tweets, analyzing their ideological perspectives. Machine learning models were applied to analyze the dynamics of sexual violence and gender justice discourses across four social media platforms.

ISBN: 9798380142151Subjects--Topical Terms:

649834
Electrical engineering.
Subjects--Index Terms:

Social media platforms

Improving Text Representations With Language Models: Contrastive Learning, Personalization, and Data Augmentation.
LDR:03099nmm a2200373 4500 001 2428045
005 20241210085348.5
006 m o d
007 cr#unu||||||||
008 261011s2023 ||||||||||||||||| ||eng d
020 $a 9798380142151
035 $a (MiAaPQ)AAI30637304
035 $a AAI30637304
040 $a MiAaPQ $c MiAaPQ
100 1 $a Lian, Ruixue. $3 3815084
245 1 0 $a Improving Text Representations With Language Models: Contrastive Learning, Personalization, and Data Augmentation.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 117 p.
500 $a Source: Dissertations Abstracts International, Volume: 85-03, Section: A.
500 $a Advisor: Sethares, William A.
502 $a Thesis (Ph.D.)--The University of Wisconsin - Madison, 2023.
520 $a Text representation learning is a fundamental task in NLP. It aims to capture the semantic, syntactic, and contextual information present in textual data and transform it into a numerical representation that machine learning models can effectively utilize. Effective text representations are crucial for various downstream tasks such as text classification, information retrieval, machine translation, and question answering, etc.This thesis presents various methodologies to improve sentence-level text representations with language models. Given the fact that subcategories under the same branches are similar and closer to each other, we propose variations of label-aware supervised contrastive loss (LA-SCL) to incorporate and learn the label hierarchy information. Two studies were conducted to improve text representations with personalized features, which can be tailored to specific users by incorporating individual information such as user demographics, or contextual factors. A novel model architecture called personalized transformer memory (PersonalTM) was proposed to effectively involve personalized features given a transformer-based encoder-decoder model on information retrieval tasks. An incremental learning algorithm was proposed to continuously keep the model up-to-date by incorporating streaming personalized data. A study was conducted to perform data augmentation by utilizing GPT-2 to generate text data, specifically focusing on producing utterances that are similar to original data in both semantics and sentiment.Several practical tasks in Social Media Analysis are also conducted in this thesis: several language models were employed to assess the favorability and hesitancy towards COVID-19 on tweets, analyzing their ideological perspectives. Machine learning models were applied to analyze the dynamics of sexual violence and gender justice discourses across four social media platforms.
590 $a School code: 0262.
650 4 $a Electrical engineering. $3 649834
650 4 $a Computer science. $3 523869
650 4 $a Mass communications. $3 3422380
653 $a Social media platforms
653 $a Contrastive learning
653 $a Transformer memory
653 $a Tweets
690 $a 0544
690 $a 0984
690 $a 0708
710 2 $a The University of Wisconsin - Madison. $b Electrical and Computer Engineering. $3 3815085
773 0 $t Dissertations Abstracts International $g 85-03A.
790 $a 0262
791 $a Ph.D.
792 $a 2023
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30637304