Shopping Basket Recommendation

Question

Data mining basket analysis we have three tabs

in third tab we have Shopping Basket Recommendation

in this tab we have last column Importance

how calculate this column i am unable to understand

any one have experience on this help me

Thanks,

T. Chinni Krishna

Answer 1

Hi T. Chinni Krishna,

The Shopping Basket Analysis tool uses the Microsoft Association Rules algorithm to detect the relationship of items frequently purchased together.

Importance is also called the interesting score (or the lift in some literature). Importance can be used to measure itemsets and rules. The importance of an itemset is defined using the formula: Importance ({A,B}) = Probability (A, B)/(Probability (A)*Probability (B))

If importance = 1, A and B are independent items. It means that the purchase of product A and the purchase of product B are two independent events. If importance < 1, A and B are negatively correlated, which means that if a customer buys A, it is unlikely he or she will also buy B. If importance > 1, A and B are positively correlated, which means that if a customer buys A, it is very likely he or she also buys B.

For rules, the importance is calculated using the formula: Importance (A => B) = log (p(B|A)/p(B|not A))

An importance of 0 means that there is no association between A and B. A positive importance score means that the probability of B goes up when A is true. A negative importance score means that the probability of B goes down when A is true.

For more details about it, please refer to: http://msdn.microsoft.com/en-us/library/cc280428.aspx

Please feel free to ask if you have any question.

Thanks,
Eileen

---

Please remember to mark the replies as answers if they help and unmark them if they provide no help. This can be beneficial to other community members reading the thread.

Answer 2

Dear Eileen.

Would you explain the difference between 'Itemset' and 'Rule'? What are they? Please use examples.

Thanks.

MasYam

Answer 3

選択したアイテム 推奨     選択されたアイテムの販売  関連付け販売  関連付け販売の割合  重要度
Fenders         Mountain Bikes        1238     539      43.54%   0.52
Hydration Packs Bottles and Cages  428     191      44.63%   0.33
Bike Stands Tires and Tubes           130     103      79.23%   0.26
Gloves             Helmets                 849     352      41.46%   0.17
Cleaners       Tires and Tubes         525     259      49.33%   0.05
Bike Racks Tires and Tubes             191      94      49.21%   0.05
Helmets         Tires and Tubes      3794   1617      42.62%  -0.02

Dear Eileen,
The above is the shopping basket result on the sample data 'associate'.
Importance (A => B) = log (p(B|A)/p(B|not A)) <---Please explain the right most col.= Importance (A => B): 0.52, 0.33, 0.26, .....
Especially how you obtain p(B|not A).

Thank you.

MasYam