how do the mining structure for a RETAIL cluster analysis..

Question

Hi 

how I can do the mining structure for a RETAIL  cluster analysis.. I have a data set that has order-detail... for example I wan the clusters with focus on  Customer has Key.. and then  nmake clusters about Customers and their details like Porduct and category..  I've tried and the results are inconsistent, any one have a idea??

Id	Order	Detail	Customer	Category	Product	Quantity	Amount
1	2	1	Cust 1	Categ 1	A	1	123
2	2	2	Cust 2	Categ 2	B	2	412
3	2	3	Cust 3	Categ 3	D	3	543
4	4	2	Cust 3	Categ 3	E	5	312
5	5	1	Cust 1	Categ 1	T	6	533
6	6	1	Cust 2	Categ 2	Y	7	134
7	6	2	Cust 3	Categ 3	F	8	123
8	8	1	Cust 3	Categ 3	S	5	143
9	8	2	Cust 1	Categ 1	W	7	321
10	9	1	Cust 2	Categ 2	R	3	234

 

---

UDA

Answer 1

Hello Franklin,

For retail market basket analysis, two algorithms that I find useful are:

• Clustering Algorithm(Cluster on Age group, Cluster on income group)
• Association Rules(If the buyer purchases, a soap then placing a shampoo and conditioners in nearby shelves might increase the order item count)

---

Please vote as helpful or mark as answer, if it helps

Answer 2

Hi Raunak

I need the structure.. I Know the algoritms and the focus but the problem is the structure..

I found an option when this work I'll share the solution. 

---

UDA

Answer 3

Hi Franklin,

Are you talking about the minig structure and yes please share your solution, perhaps we can fine tune the model.

---

Please vote as helpful or mark as answer, if it helps