0.25
0.5
0.75
1.25
1.5
1.75
2
Finding frequent patterns from data
Published on Feb 25, 20079256 Views
Discovery of frequent patterns = finding positive conjunctions that are true for a given fraction of the observations - this basic idea can be instantiated in many ways: - finding frequent sets from 0
Related categories
Chapter list
Finding frequent patterns from data00:01
Main themes01:50
Main themes (cont.)03:50
Chapter 1: Introduction05:50
What is data mining?05:59
Example: document collections07:36
Example: document collections09:12
Example: document collections09:32
Example: telecommunications alarm data09:55
Example: web page link structure11:08
Properties of the data sets12:03
Example: web page link structure13:44
Example: document collections14:00
Properties of the data sets14:20
Application areas14:35
Theory and practice15:20
Two traditions in data mining16:56
Local and global modeling17:44
Fast counting18:42
Chapter 2: Frequent sets and association rules19:23
Frequent sets and ssociation rules19:38
Example20:02
Properties of the data20:56
Association rules21:40
Example dataset23:15
Basic statistics23:28
What is there in the data?25:18
How often a variable is 1?25:27
How often a variable is 1?26:08
How often a variable is 1?26:19
Number of ones per row26:44
Number of ones per row27:46
Number of rows with a given number of ones28:20
Correlations between variables28:57
Correlations between variables30:15
Histogram of correlations31:07
Frequent sets and rules from the retail data set31:19
Association rules31:45
Problem formulation: data32:16
Problem formulation: data33:05
Notation34:04
Patterns: sets of items35:17
Problem formulation: data35:48
Patterns: sets of items36:15
Example36:44
Frequent sets37:16
Example37:50
Frequent sets38:45
Association rules39:04
Association rules II39:44
Discovery task40:25
Discovery task41:39
Example41:54
Discovery task42:24
How to find association rules43:03
Histogram of correlations43:12
Rules with threshold 100043:21
Interestingness of rules44:13
Rules with threshold 100044:19
Interestingness of rules44:34
Rules with threshold 100045:10
How to find association rules46:18
Finding frequent sets47:55
Candidates of size l+1? Monotonicity!49:43
Candidates of size l+1? Monotonicity!50:48
Example51:59
Apriori algorithm for frequent sets53:44
Correctness54:57
Candidate generation55:05
Apriori algorithm for frequent sets55:15
Candidate generation56:00
Candidate generation algorithm57:34
Correctness and running time57:36
Database pass58:00
Apriori algorithm for frequent sets58:10
Database pass58:25
Correctness59:35
Data structures01:00:39
Algorithm01:01:08
Correctness01:03:21
Apriori algorithm for frequent sets01:03:29
Implementation details01:04:02
Example optimization with little effect on efficiency01:04:38
More theoretical approaches01:05:37
Are there useless candidates?01:06:44
Additional information can change things...01:08:07
A very simple lower bound01:10:50
Another very simple lower bound01:11:35
Theoretical analyses01:11:59
Random databases01:12:16
Random databases01:13:10
Random databases01:13:24
Number of frequent sets01:13:54
Sampling for finding association rules01:15:10
Extensions01:15:50
What does one do with the rules01:16:45
Frequent sets, again01:17:11
Finding frequent patterns01:18:29
Generalization, cont.01:20:07
Chapter 3. Another example: episodes01:20:58