A Unified Understanding of Deep NLP Models for Text Classification

The rapid development of deep natural language processing (NLP) models for text classification has led to an urgent need for a unified understanding of these models proposed individually. Existing methods cannot meet the need for understanding different models in one framework due to the lack of a u...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on visualization and computer graphics. - 1996. - 28(2022), 12 vom: 01. Dez., Seite 4980-4994
1. Verfasser: Li, Zhen (VerfasserIn)
Weitere Verfasser: Wang, Xiting, Yang, Weikai, Wu, Jing, Zhang, Zhengyan, Liu, Zhiyuan, Sun, Maosong, Zhang, Hui, Liu, Shixia
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2022
Zugriff auf das übergeordnete Werk:IEEE transactions on visualization and computer graphics
Schlagworte:Journal Article Research Support, Non-U.S. Gov't
LEADER 01000caa a22002652c 4500
001 NLM342455192
003 DE-627
005 20250303115658.0
007 cr uuu---uuuuu
008 231226s2022 xx |||||o 00| ||eng c
024 7 |a 10.1109/TVCG.2022.3184186  |2 doi 
028 5 2 |a pubmed25n1141.xml 
035 |a (DE-627)NLM342455192 
035 |a (NLM)35724276 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Li, Zhen  |e verfasserin  |4 aut 
245 1 2 |a A Unified Understanding of Deep NLP Models for Text Classification 
264 1 |c 2022 
336 |a Text  |b txt  |2 rdacontent 
337 |a ƒaComputermedien  |b c  |2 rdamedia 
338 |a ƒa Online-Ressource  |b cr  |2 rdacarrier 
500 |a Date Completed 28.10.2022 
500 |a Date Revised 15.11.2022 
500 |a published: Print-Electronic 
500 |a Citation Status MEDLINE 
520 |a The rapid development of deep natural language processing (NLP) models for text classification has led to an urgent need for a unified understanding of these models proposed individually. Existing methods cannot meet the need for understanding different models in one framework due to the lack of a unified measure for explaining both low-level (e.g., words) and high-level (e.g., phrases) features. We have developed a visual analysis tool, DeepNLPVis, to enable a unified understanding of NLP models for text classification. The key idea is a mutual information-based measure, which provides quantitative explanations on how each layer of a model maintains the information of input words in a sample. We model the intra- and inter-word information at each layer measuring the importance of a word to the final prediction as well as the relationships between words, such as the formation of phrases. A multi-level visualization, which consists of a corpus-level, a sample-level, and a word-level visualization, supports the analysis from the overall training set to individual samples. Two case studies on classification tasks and comparison between models demonstrate that DeepNLPVis can help users effectively identify potential problems caused by samples and model architectures and then make informed improvements 
650 4 |a Journal Article 
650 4 |a Research Support, Non-U.S. Gov't 
700 1 |a Wang, Xiting  |e verfasserin  |4 aut 
700 1 |a Yang, Weikai  |e verfasserin  |4 aut 
700 1 |a Wu, Jing  |e verfasserin  |4 aut 
700 1 |a Zhang, Zhengyan  |e verfasserin  |4 aut 
700 1 |a Liu, Zhiyuan  |e verfasserin  |4 aut 
700 1 |a Sun, Maosong  |e verfasserin  |4 aut 
700 1 |a Zhang, Hui  |e verfasserin  |4 aut 
700 1 |a Liu, Shixia  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE transactions on visualization and computer graphics  |d 1996  |g 28(2022), 12 vom: 01. Dez., Seite 4980-4994  |w (DE-627)NLM098269445  |x 1941-0506  |7 nnas 
773 1 8 |g volume:28  |g year:2022  |g number:12  |g day:01  |g month:12  |g pages:4980-4994 
856 4 0 |u http://dx.doi.org/10.1109/TVCG.2022.3184186  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d 28  |j 2022  |e 12  |b 01  |c 12  |h 4980-4994