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

Natural Language Processing for Health System Messages: Deep Transfer Learning Approach to Aspect-Based Sentiment Analysis of COVID-19 Content.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Natural Language Processing for Health System Messages: Deep Transfer Learning Approach to Aspect-Based Sentiment Analysis of COVID-19 Content./
作者:	Sun, Mary.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2022,
面頁冊數:	109 p.
附註:	Source: Masters Abstracts International, Volume: 83-08.
Contained By:	Masters Abstracts International83-08.
標題:	Computer science. -
ISBN:	9798790637230

Natural Language Processing for Health System Messages: Deep Transfer Learning Approach to Aspect-Based Sentiment Analysis of COVID-19 Content.
Sun, Mary.

Natural Language Processing for Health System Messages: Deep Transfer Learning Approach to Aspect-Based Sentiment Analysis of COVID-19 Content. - Ann Arbor : ProQuest Dissertations & Theses, 2022 - 109 p.

Source: Masters Abstracts International, Volume: 83-08.

Thesis (A.L.M.)--Harvard University, 2022.

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

Recent efforts in clinical natural language processing research have focused on mining social media data and other user-generated data relevant to COVID-19. An increasing number of machine learning tools have been developed for annotation, topic modeling, and various classification tasks, including the now widely used Bidirectional Encoder Representations from Transformer (BERT) language model. In this study, we improve the domain specificity of COVID-Twitter-BERT for secure health system messages through supervised fine-tuning on a large corpus of patient-originated messages. Following sentiment annotation with a validated rule-based heuristic tool, an auxiliary sentence approach was implemented to conduct aspect-based sentiment analysis. We find that existing machine learning tools optimized on social media tools demonstrate moderate-to-good generalizability to secure patient messages. Furthermore, significant performance improvements to a weighted average F1 score of 0.881 were realized upon implementation of fine-tuning and auxiliary sentence approaches. We also observe temporal trends in positive, negative, and compound (overall) sentiment measured on over 3.2 million patient messages related to COVID-19. We believe that secure health system messages are an increasingly valuable source of clinical text data and can be effectively mined using emerging deep transfer learning techniques.

ISBN: 9798790637230Subjects--Topical Terms:

523869
Computer science.
Subjects--Index Terms:

Clinical informatics

Natural Language Processing for Health System Messages: Deep Transfer Learning Approach to Aspect-Based Sentiment Analysis of COVID-19 Content.
LDR:02686nmm a2200409 4500 001 2349686
005 20221003074930.5
008 241004s2022 eng d
020 $a 9798790637230
035 $a (MiAaPQ)AAI28965422
035 $a AAI28965422
040 $a MiAaPQ $c MiAaPQ
100 1 $a Sun, Mary. $0 (orcid)0000-0003-3964-3858 $3 3689102
245 1 0 $a Natural Language Processing for Health System Messages: Deep Transfer Learning Approach to Aspect-Based Sentiment Analysis of COVID-19 Content.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2022
300 $a 109 p.
500 $a Source: Masters Abstracts International, Volume: 83-08.
500 $a Advisor: Takeshita, Junko;Wang, Hongming.
502 $a Thesis (A.L.M.)--Harvard University, 2022.
506 $a This item must not be sold to any third party vendors.
520 $a Recent efforts in clinical natural language processing research have focused on mining social media data and other user-generated data relevant to COVID-19. An increasing number of machine learning tools have been developed for annotation, topic modeling, and various classification tasks, including the now widely used Bidirectional Encoder Representations from Transformer (BERT) language model. In this study, we improve the domain specificity of COVID-Twitter-BERT for secure health system messages through supervised fine-tuning on a large corpus of patient-originated messages. Following sentiment annotation with a validated rule-based heuristic tool, an auxiliary sentence approach was implemented to conduct aspect-based sentiment analysis. We find that existing machine learning tools optimized on social media tools demonstrate moderate-to-good generalizability to secure patient messages. Furthermore, significant performance improvements to a weighted average F1 score of 0.881 were realized upon implementation of fine-tuning and auxiliary sentence approaches. We also observe temporal trends in positive, negative, and compound (overall) sentiment measured on over 3.2 million patient messages related to COVID-19. We believe that secure health system messages are an increasingly valuable source of clinical text data and can be effectively mined using emerging deep transfer learning techniques.
590 $a School code: 0084.
650 4 $a Computer science. $3 523869
650 4 $a Artificial intelligence. $3 516317
650 4 $a Medicine. $3 641104
650 4 $a Public health. $3 534748
650 4 $a Bioinformatics. $3 553671
653 $a Clinical informatics
653 $a COVID-19
653 $a Health system messages
653 $a Large-scale data mining
653 $a Natural language processing
653 $a Sentiment analysis
653 $a Coronavirus disease 2019
690 $a 0984
690 $a 0800
690 $a 0564
690 $a 0715
690 $a 0573
710 2 0 $a Harvard University. $b Extension Studies. $3 3560717
773 0 $t Masters Abstracts International $g 83-08.
790 $a 0084
791 $a A.L.M.
792 $a 2022
793 $a English