en
0.25
0.5
0.75
1.25
1.5
1.75
2
Declarative Modeling For Machine Learning and Data Mining
Published on Oct 29, 20124660 Views
Despite the popularity of machine learning and data mining today, it remains challenging to develop applications and software that incorporates machine learning or data mining techniques. This is beca
Related categories
Chapter list
Declarative Modeling for Machine Learning and Data Mining00:00
Our work today ... - 100:20
And do it again ...01:00
Our work today ... - 201:17
The Challenge01:34
Key Point01:43
Overview Talk02:35
The What, Why and How of Declarative Modeling03:05
What is declarative modeling?03:37
How does it work? - 105:18
Why declarative modeling?06:20
Constraint Programming07:41
Constraint Satisfaction08:29
Solvers for CP09:21
Search - 111:27
Search - 211:31
Search - 311:44
Search - 411:54
Search - 511:59
Search - 612:21
Search - 712:29
Search - 812:32
Search - 912:38
Search - 1012:38
Search - 1112:40
Constraint Programming12:41
What about ML/DM?14:05
Observation 1 - a14:18
Data Mining14:32
Itemset mining15:06
Machine learning15:20
Observation 1 - b15:55
Observation 2 - a16:13
Observation 2 - b17:06
Observation 317:54
Long standing open questions18:36
Questions remain open19:29
Evidence: The case of Pattern mining20:35
Pattern Mining - 120:46
Pattern Mining - 222:10
Frequent Pattern Mining23:10
Frequent Itemset Mining - 123:13
Frequent Itemset Mining - 224:05
Frequent Itemset Mining - 325:06
Closed Itemset Mining26:00
Further Constraints26:46
How does it work? - 227:06
Solver 127:22
Encoding in Zinc27:46
Resulting Search Strategy akin to Zaki’s Eclat [KDD 97]28:30
Solver 228:50
Correlated Pattern Mining29:39
Top-k Correlated Pattern Mining Subgroup Discovery29:44
Modeling perspective29:54
Correlation function30:28
Monotonicity31:10
Illustration - 132:29
Illustration - 233:26
Illustration - 334:06
Illustration - 434:07
4-support bound34:09
2-support bound34:35
1-support bound34:39
Experiments34:47
Pattern Set Mining35:05
Pattern Sets - 135:15
Pattern Sets - 235:26
Pattern Sets - 336:15
k-Pattern Set Mining (|P|=k)36:47
Generality37:01
Pattern Mining37:32
http://dtai.cs.kuleuven.be/CP4IM38:05
Perspective38:17
All this is fine but...38:27
Efficiency / Scalability38:46
Task / Representation40:28
kLOG [Frasconi, Costa, DR, De Grave 12]41:39
A biomedical NLP task [Verbeke et al. EMLNP 12]42:22
E/R-MODEL42:41
Relational ....43:21
Graphicalization43:31
kLOG - 143:47
kLOG - 243:51
NSPDK [Costa et al ICML 10]44:01
kLOG - 344:28
kLOG - 545:00
What if we succeed?45:31
Our work today ... - 345:49
Our work tomorrow ... - 146:07
Our work tomorrow ... - 247:09
Conclusions - 147:49
Conclusions - 248:38
Questions?48:53