Association Rule Mining (Data Mining Tool) COMP1942 Association Rule Mining (Data Mining Tool) Prepared by Raymond Wong (Notes) and Kai Ho Chan (Screenshot) Presented by Raymond Wong raywong@cse COMP1942
Outline Association Rule Mining Algorithm Problem Definition Algorithm How to use the data mining tool COMP1942
How to use the data mining tool Where can I find the data mining tool? How can I use the data mining tool for association rule mining? COMP1942
Where can I find the data mining tool? There are two ways of opening XLMiner in MS Excel. Option 1: From the “Add-ins” Tag Option 2: From the “XLMiner Platform” Tag Suggestion: Use the “Add-ins” Tag COMP1942
COMP1942
COMP1942
COMP1942
COMP1942
COMP1942
How to add “XLMiner” in MS Excel installed in a PC of our lab There are two ways of opening XLMiner in MS Excel. Option 1: From the “Add-ins” Tag Option 2: From the “XLMiner Platform” Tag COMP1942
COMP1942
COMP1942
COMP1942
COMP1942
How to use the data mining tool Where can I find the data mining tool? How can I use the data mining tool for association rule mining? COMP1942
Where can I find the data mining tool? Open “rule.xls” in MS Excel COMP1942
COMP1942
COMP1942
COMP1942
COMP1942
Data source Workbook Worksheet Data range COMP1942
COMP1942
COMP1942
# Rows in data 5 5 # Columns in data COMP1942
First row contains header COMP1942
Data in binary matrix format Input data format Data in binary matrix format Data in item list format COMP1942
Minimum support (# transactions): Parameters Minimum support (# transactions): 3 Minimum confidence (%): COMP1942
Minimum support (# transactions): Parameters Minimum support (# transactions): 3 50 Minimum confidence (%): COMP1942
COMP1942
COMP1942
COMP1942
Antecedent (A) Row ID Support for A Consequent (C) Conf. % Support for C Support for A & C Lift Ratio COMP1942
Rule 1: If item D is purchased, then this implies item A is also purchased. This rule has confidence of 100% COMP1942
Rule 5: If item B is purchased, then this implies item C is also purchased. This rule has confidence of 75% COMP1942
In the previous setting, we set Minimum Support = 3 Minimum Confidence = 50% In the following setting, Minimum Support = 2 Minimum Confidence = 70% COMP1942
COMP1942
Minimum support (# transactions): Parameters Minimum support (# transactions): 2 70 Minimum confidence (%): COMP1942
COMP1942
Rule 1: If items D, E are purchased, then this implies items A, B are also purchased. This rule has confidence of 100% COMP1942