東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Scaling NLP Models to Analyze Large-...

Babvey, Pouria,

Linked to FindBook

Google Book

Amazon

博客來

Scaling NLP Models to Analyze Large-Scale Online Conversations /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Scaling NLP Models to Analyze Large-Scale Online Conversations // Pouria Babvey.
Author:	Babvey, Pouria,
Description:	1 electronic resource (159 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-01, Section: B.
Contained By:	Dissertations Abstracts International86-01B.
Subject:	Systems science. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31240737
ISBN:	9798383181461

Scaling NLP Models to Analyze Large-Scale Online Conversations /
Babvey, Pouria,

Scaling NLP Models to Analyze Large-Scale Online Conversations /Pouria Babvey. - 1 electronic resource (159 pages)

Source: Dissertations Abstracts International, Volume: 86-01, Section: B.

In the past decade, Artificial Intelligence (AI) has experienced a notable renaissance. AI's capabilities in parsing complex systems have been particularly evidenced in the advances within Natural Language Processing (NLP). The introduction of the Word2vec algorithm in 2013 offered a means to translate text into numerical vectors and catalyzed the application of convolutional and recurrent neural networks to text analysis. In 2017, Transformers revolutionized the NLP field and started a new chapter in AI with large language models.Although large language models are highly skilled at answering questions and can access a broad range of knowledge, there remains a need for methods that can visually capture the extent and dynamics of conversations on social media. The applications of AI to complex systems such as large-scale online conversations still present several challenges. Online conversations represent a complex and dynamic fabric of societal discourse with its unique contextual cues and subtleties that are distributed and constantly evolving with no specific boundaries and unpredictable outcomes. Our research has been built around this mission. By marrying the foundational concept of "frames" introduced by Minsky with contemporary AI methodologies, our findings delve into the intricacies of online dialogues, unveiling the deficiencies of current models in capturing the holistic context required for tasks like stance detection. The research aims to develop new tools to analyze online conversations, how they form, the spread and influence of different types of information via these conversations, how the conversations are cross-cutting or coherent, what are the target topics, and how participant opinions toward the target are distributed, and if the extracted knowledge can help NLP models to obtain a better understanding of the context of discussion subjects. The computational tools and data analytics techniques developed in our studies complement the previous research in mining online conversations and providea useful framework to better understand large-scale conversations. Our developed methods have been used in several studies by other researchers for hate speech detection, disaster risk reduction, and social and political opinion mining paving the way for more advanced and contextually aware AI systems.

English

ISBN: 9798383181461Subjects--Topical Terms:

3168411
Systems science.
Subjects--Index Terms:

Disaster risk reduction

Scaling NLP Models to Analyze Large-Scale Online Conversations /
LDR:03913nmm a22004693i 4500 001 2431248
005 20250603111413.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798383181461
035 $a (MiAaPQD)AAI31240737
035 $a AAI31240737
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Babvey, Pouria, $e author. $3 3819964
245 1 0 $a Scaling NLP Models to Analyze Large-Scale Online Conversations / $c Pouria Babvey.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (159 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-01, Section: B.
500 $a Advisors: Lipizzi, Carlo; Ramirez-Marquez, Jose Committee members: Iandoli, Luca; Kowal, Michael.
502 $b Ph.D. $c Stevens Institute of Technology $d 2024.
520 $a In the past decade, Artificial Intelligence (AI) has experienced a notable renaissance. AI's capabilities in parsing complex systems have been particularly evidenced in the advances within Natural Language Processing (NLP). The introduction of the Word2vec algorithm in 2013 offered a means to translate text into numerical vectors and catalyzed the application of convolutional and recurrent neural networks to text analysis. In 2017, Transformers revolutionized the NLP field and started a new chapter in AI with large language models.Although large language models are highly skilled at answering questions and can access a broad range of knowledge, there remains a need for methods that can visually capture the extent and dynamics of conversations on social media. The applications of AI to complex systems such as large-scale online conversations still present several challenges. Online conversations represent a complex and dynamic fabric of societal discourse with its unique contextual cues and subtleties that are distributed and constantly evolving with no specific boundaries and unpredictable outcomes. Our research has been built around this mission. By marrying the foundational concept of "frames" introduced by Minsky with contemporary AI methodologies, our findings delve into the intricacies of online dialogues, unveiling the deficiencies of current models in capturing the holistic context required for tasks like stance detection. The research aims to develop new tools to analyze online conversations, how they form, the spread and influence of different types of information via these conversations, how the conversations are cross-cutting or coherent, what are the target topics, and how participant opinions toward the target are distributed, and if the extracted knowledge can help NLP models to obtain a better understanding of the context of discussion subjects. The computational tools and data analytics techniques developed in our studies complement the previous research in mining online conversations and providea useful framework to better understand large-scale conversations. Our developed methods have been used in several studies by other researchers for hate speech detection, disaster risk reduction, and social and political opinion mining paving the way for more advanced and contextually aware AI systems.
546 $a English
590 $a School code: 0733
650 4 $a Systems science. $3 3168411
650 4 $a Information science. $3 554358
650 4 $a Computer science. $3 523869
650 4 $a Web studies. $3 2122754
653 $a Disaster risk reduction
653 $a Natural Language Processing
653 $a Social landscapes
653 $a Social network analysis
653 $a Transformers
690 $a 0790
690 $a 0723
690 $a 0984
690 $a 0646
690 $a 0800
710 2 $a Stevens Institute of Technology. $b School of Systems and Enterprises. $e degree granting institution. $3 3819935
720 1 $a Lipizzi, Carlo $e degree supervisor.
720 1 $a Ramirez-Marquez, Jose $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-01B.
790 $a 0733
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31240737