I need help with creating an Alteryx project to
transform an edge list like an example below (see in Input tab of the attached file):
Account | TransactionID |
3000003Ct | 20134000000001 |
3000017Dt | 20134000000001 |
3301102Dt | 20134000000001 |
3000003Ct | 20134000000002 |
3000016Dt | 20134000000002 |
3301102Dt | 20134000000002 |
3000003Ct | 20134000000003 |
3000016Dt | 20134000000003 |
3301102Dt | 20134000000003 |
into adjacency matrix like the one below (see in Output tab of the attached file):
| 3000003Ct | 3000017Dt | 3301102Dt | 3000016Dt |
3000003Ct | 0 | 1 | 3 | 2 |
3000017Dt | 1 | 0 | 1 | 0 |
3301102Dt | 3 | 1 | 0 | 2 |
3000016Dt | 2 | 0 | 2 | 0 |
Thank you
Solved! Go to Solution.
Hi @EugeneY
See this solution:
- Use Summarize Tool to group by Transactions and Concatenate all Accounts
- Join the original dataset to the concatenated accounts for each Transaction ID (join by Transaction ID)
- Use Text to Columns to Split to Rows, so you have accounts connected in a row level
- Group by and count each relationship between accounts
- If you are comparing same accounts, that's gotta count as a 0, so we use formula tool.
- Cross-Tab Tool to set up the matrix
- Data Cleansing Tool to transform Nulls in 0's.
Workflow attached on version 2018.4
Cheers,
Thank you, Thableaus.
Your code produces expected results.
I appreciate your help.
User | Count |
---|---|
106 | |
82 | |
70 | |
54 | |
40 |