Modeling and Summarizing News Events using Semantic Triples thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Modeling and Summarizing News Events using Semantic Triples

Published on Jul 10, 2018663 Views

Summarizing news articles is becoming crucial for allowing quick and concise access to information about daily events. This task can be challenging when the same event is reported with various levels

Related categories

Chapter list

Modeling and Summarizing News Events Using Semantic Triples00:00
Eve the News Reader00:22
Wall-e promises…00:42
Extractive vs Abstractive00:45
The Chemistry of Abstractive Summarization02:01
Abstractive News Summarization Approaches03:14
PSB Example03:51
PSB Limitation04:42
PSB Limitation: Subclause/Adverbial Fact04:53
Pattern Graph Fusion (PGF)05:47
Pattern Graph Fusion: Extraction07:21
Pattern Graph Fusion: Typing07:43
Fusion of Sentences along Types08:04
Pattern Graph Fusion: Limitations 08:34
Pattern Graph Fusion: Limitations 09:01
Our Summarization09:28
Our Summarization: Pipeline09:49
Pipeline: Entity Linking10:31
Pipeline: Verb Linking11:33
Pipeline: Fusion and Ranking - 111:45
Pipeline: Fusion and Ranking - 212:42
Pipeline: Summary Generation - 114:07
Pipeline: Summary Generation - 214:32
Evaluation15:21
Results: Rouge16:08
Results: Pyramid16:43
Conclusions17:16
Future work17:51
Thank you!18:36