en
0.25
0.5
0.75
1.25
1.5
1.75
2
The Battle for the Future of Data Mining
Published on 2014-10-079821 Views
Deep learning has catapulted to the front page of the New York Times, formed the core of the so-called 'Google brain', and achieved impressive results in vision, speech recognition, and elsewhere. Yet
Related categories
Presentation
The Battle for the Future of Data Mining00:00
Big Data Tidal Wave00:49
What's Next?01:21
Deep Learning01:27
Some Skepticism about Deep Learning01:33
Outline02:09
What are the limits of Classification and prediction?03:18
Example of State of the Art 03:39
Product Recommendations…05:26
Article08:00
More Challenges for Classification08:30
Fundamental Claim10:58
The Bottom Line12:17
Winograd Schema Challenge (Levesque, 2011)12:56
Just the Facts…14:20
2014 Knowledge Tour Google15:43
WolframAlpha15:47
OpenCyc16:03
ConceptNet16:12
Observations from the Tour16:35
III. Machine Reading17:15
Information Extraction (IE)17:22
NELL Lexical Patterns for “haswife”18:34
NELL Knowledge Base19:28
Open Information Extraction (2007)19:51
History of Open IE21:32
ReVerb22:13
IV. The Allen Institute for AI (AI2)22:45
Mission + Talent = Impact23:37
AI2 Overview (Nonprofit Research Institute)24:25
Flagship Project: Aristo25:16
Aristo: 4th, 5th, 6th … Grade Science Tests26:29
Factual Knowledge for 4th Grade Science27:08
Summary of our Approach28:02
Geometry Domain (Case Study)29:05
Geometry Diagram Alone30:14
Geometry Text Alone30:17
Need Both Diagram and Text30:22
Diagram Understanding30:25
Experimental Study31:13
“Semantic Scholar” Project31:53
Ambitious Research Investing in Engineering Mindset33:46
Summary35:26
Thank you from AI2!36:36
What is Data Mining?36:45