利用 Attentive 來改善端對端中文語篇剖析遞迴類 神經網路系統;Using Attentive to improve Recursive LSTM End-to- End Chinese Discourse Parsing

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/81262

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/81262

Title:	利用 Attentive 來改善端對端中文語篇剖析遞迴類神經網路系統;Using Attentive to improve Recursive LSTM End-to- End Chinese Discourse Parsing
Authors:	王育任;Wang, Yu-Jen
Contributors:	資訊工程學系
Keywords:	深度學習;篇章剖析;注意力機制;遞迴類神經網路;Deep Learning;Discourse Parsing;Attention;Recursive neural network
Date:	2019-08-08
Issue Date:	2019-09-03 15:41:10 (UTC+8)
Publisher:	國立中央大學
Abstract:	篇章剖析，可以幫助我們以不同角度來理解文句之間的關係與連結，但篇章剖析的資料結構目前仰賴人工標記，使得這項技術無法直接被利用在任意篇章中。因此至目前為止，有許多研究著手於讓電腦能夠自動對篇章進行剖析，並建立出一個完整的剖析樹。以中文語料庫 CDTB 來說，欲建立完整的篇章剖析程式，其問題主要可以被分成四個，分別是子句分割、剖析樹建立、子句關係辨識、中心關係辨識。由於深度學習近幾年發展快速，因此針對篇章剖析的建構方法也從傳統的 SVM, CRF 等，進展到目前已遞迴類神經方式來建構剖析篇章。而在本篇論文中，我們也加入了許多目前最新的深度學習技術，例如 Attentive RvNN、self-attentive、BERT 等方法，來提高模型的準確度。最後，我們成功將每一項任務的 F1 都提高了近 10% 左右，達到目前我們所知研究中最好的效能。;Discourse parser can help us to understand the relationship and connection between sentences from different angles, but the tree structure data still need to rely on manual marking, which makes this technology cannot be directly used in life. So far, there have been many research studies on automatically construct the complete tree structure on the computer. Since deep learning has progressed rapidly in recent years, the construction method for discourse parser has also changed from the traditional SVM, CRF method to the current recursive neural. In the Chinese corpus tree library CDTB, the parsing analysis problem can be divided into four main problems, including elementary discourse unit (EDU) segmentation, tree structure construction, center labeling, and sense labeling. In this paper, we use many state-of-the-art deep learning techniques, such as attentive recursive neural networks, self-attentive, and BERT to improve the performance. In the end, we succeed to increase the accuracy by more than 10% of F1 of each task, reaching the best performance we know so far.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	150	View/Open

社群 sharing

Loading...