This page is an example of using TablePress for your project.

Chapter 2 in the book uses K-means clustering to cluster deals sent out by a wine dealer. Here are some results obtained on using 4 clusters-

Offer #	Offer date	Product	Minimum Qty (kg)	Discount	Origin	Past Peak	1	2	3	4
1	January	Malbec	72	56	France	FALSE	0	0	4	6
2	January	Pinot Noir	72	17	France	FALSE	4	0	4	2
3	February	Espumante	144	32	Oregon	TRUE	0	0	2	4
4	February	Champagne	72	48	France	TRUE	0	0	7	5
5	February	Cabernet Sauvignon	144	44	New Zealand	TRUE	0	0	2	2
6	March	Prosecco	144	86	Chile	FALSE	0	0	5	7
7	March	Prosecco	6	40	Australia	TRUE	0	12	4	3
8	March	Espumante	6	45	South Africa	FALSE	0	11	6	3
9	April	Chardonnay	144	57	Chile	FALSE	0	0	7	3
10	April	Prosecco	72	52	California	FALSE	0	0	5	2
11	May	Champagne	72	85	France	FALSE	0	0	7	6
12	May	Prosecco	72	83	Australia	FALSE	0	0	3	2
13	May	Merlot	6	43	Chile	FALSE	0	6	0	0
14	June	Merlot	72	64	Chile	FALSE	0	0	5	4
15	June	Cabernet Sauvignon	144	19	Italy	FALSE	0	0	2	4
16	June	Merlot	72	88	California	FALSE	0	0	5	0
17	July	Pinot Noir	12	47	Germany	FALSE	7	0	0	0
18	July	Espumante	6	50	Oregon	FALSE	0	11	2	1
19	July	Champagne	12	66	Germany	FALSE	0	0	2	3
20	August	Cabernet Sauvignon	72	82	Italy	FALSE	0	0	4	2
21	August	Champagne	12	50	California	FALSE	0	0	2	2
22	August	Champagne	72	63	France	FALSE	0	0	0	21
23	September	Chardonnay	144	39	South Africa	FALSE	0	0	3	2
24	September	Pinot Noir	6	34	Italy	FALSE	12	0	0	0
25	October	Cabernet Sauvignon	72	59	Oregon	TRUE	0	0	3	3
26	October	Pinot Noir	144	83	Australia	FALSE	8	0	5	2
27	October	Champagne	72	88	New Zealand	FALSE	0	0	6	3
28	November	Cabernet Sauvignon	12	56	France	TRUE	0	0	4	2
29	November	Pinot Grigio	6	87	France	FALSE	0	15	2	0
30	December	Malbec	6	54	France	FALSE	0	16	2	4
31	December	Champagne	72	89	France	FALSE	0	0	10	7
32	December	Cabernet Sauvignon	72	45	Germany	TRUE	0	0	3	1

 

The result obtained for different clusters can be visualized.

Cluster 1 has clear distinctions on what deals were chosen by customers. In this case the offer on Pinot Noir-

For cluster 2, some of the deals were chosen by people who could be classified as low income buyers-

Cluster 3 is not clear and has a lot of noise-

Cluster 4 shows a group of people really loved August Champaign deal. This result is quite inconclusive-