Pascal Challenge on Evaluating Machine Learning for Information Extraction: Goals, Results and Conclusions thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Pascal Challenge on Evaluating Machine Learning for Information Extraction: Goals, Results and Conclusions

Published on Feb 25, 20073767 Views

Related categories

Chapter list

PASCAL CHALLENGE ON INFORMATION EXTRACTION & MACHINE LEARNING00:00
Organisers00:04
Outline00:10
Goal : Provide a testbed for comparative evaluation of ML-based IE01:01
Data (Workshop CFP)01:53
Data (Workshop CFP)03:42
Data (Workshop CFP)04:27
Data (Workshop CFP)04:46
Preprocessing05:45
Annotation Exercise06:00
Annotation Document - sample07:00
Annotation Slots09:47
Evaluation Tasks11:01
Evaluation12:11
Participants12:33
Task113:52
Task1: Test Corpus13:55
Task1: Test Corpus14:09
Task1: 4-Fold Cross-validation15:39
Task1: 4-Fold & Test Corpus16:23
Task1: Slot FMeasure17:41
Best Slot FMeasures Task1: Test Corpus19:51
Task 2a20:46
Task2a: Learning Curve FMeasure20:53
Task2a: Learning Curve Precision22:34
Task2a: Learning Curve Recall23:00
Task 2b23:13
Active Learning (1)23:19
Active Learning (1)23:49
Active Learning (2)24:09
Active Learning (2)24:41
Active Learning (3)24:55
Active Learning (3)25:02
Task2b: Active Learning25:48
Task2b: Active Learning Increased FMeasure over random selection26:43
Task 328:27
Conclusions (Task1)28:43
Conclusions (Task1: Test Corpus)29:49
Conclusions (Task1: 4-fold Corpus)30:19
Conclusions (Task1)31:01
Conclusion (Task2 & Task3)31:40
Discussion32:25