Knimi Data Mining modeling and Analysis series _002_ using Knimi to do business Super Retail Association recommendation

Using Knimi to make business Super Retail Association recommendation

Sinom

20150801

Http://blog.csdn.net/shuaihj

First, test data

Need to test the data, please leave the mailbox

Second, the Training Association recommendation rule

1. Read the sales record (sales.table)

2. Training Association Rules ( get before and after items)

Set the Min itemsets property

3. Filter does not care about the column

only "latter" and "preceding" two columns are reserved

4. Rename name more readable

5. Data Flow

6. Training Results

Iii. Aggregation of latter information

1. Read product information (items.table)

2. Aggregate latter details

Set aggregation methods and key fields

3. Data Flow

4. Aggregation Results

Iv. aggregation of the preceding information

1. Copy the row ID ( to provide a basis for the fit)

Take a new name for the Copy column

Copy the result:

2. Remove the preceding paragraph

Splitting results

3. Aggregate the preceding paragraph for more information

Set aggregation methods and key fields

4. Rename name more readable

5. Merging latter

Set merge by Key field

after you set up the merge, the " preceding name" is grouped together

6. Merge Results

7. Data Flow

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

Knimi Data Mining modeling and Analysis series _002_ using Knimi to do business Super Retail Association recommendation