Webclass 94: Did you know that you could use a simple python program to flag unusual payment transactions?

Q1. Why is it maybe not enough to look at bank payment transactions from the ERP system?

Because the bank account details can be changed after the bank payments file has left the system

Correct Incorrect Correct answer

Because changes can be made within the ERP system that would update the bank file

Correct Incorrect Correct answer

Q2. How does Isolation Forest determine that something is unusual?

Unusual items are those for which we need to ask the least questions to separate them from the rest of the population

Correct Incorrect Correct answer

Unusual items are those that have the highest value

Correct Incorrect Correct answer

Q3. Does Isolation Forest work on one field value or many field values?

Many field values, it considers all of the values in the different fields together

Correct Incorrect Correct answer

One field value, especially that relating to value

Correct Incorrect Correct answer

Q4. If someone did many transactions of similar value just below a threshold approval rating over many years, would Isolation Forest help to pick up the fraud?

No, Benford's law would be more likely to pick that up

Correct Incorrect Correct answer

Yes, Isolation Forest would be able to pick it up

Correct Incorrect Correct answer

Q5. From the REGUH table, would we want to include the payment accounting document number (VBLNR) in the Isolation Forest analysis?

No, because it is an internal system number

Correct Incorrect Correct answer

Yes

Correct Incorrect Correct answer

Q6. What would be more interesting to include, other than dates?

The difference between the posting date and the invoice date (that we could get via REGUP)

Correct Incorrect Correct answer

The year of the posting date

Correct Incorrect Correct answer

Q7. Which field in the REGUH table helps us to know if the payment table was changed?

XAEND

Correct Incorrect Correct answer

XVORL

Correct Incorrect Correct answer

Q8. Should we include all of the records from the settlements data table (REGUH)?

No, we should not include those that have XVORL = X

Correct Incorrect Correct answer

Yes, we should include all records in order to be able to have a fair and unbiased analysis

Correct Incorrect Correct answer

Q9. Why might an anomaly such as an unusual bank account for a third-party not be picked up by a “normal” audit test that does not use artificial intelligence?

Because it depends on the rules that are applied, for example, if we say that a bank account is unusual for a third-party if it is found only 1/100 times or less, but there might be an anomaly for a bank account found only 1/50 times which we would then miss

Correct Incorrect Correct answer

Because in a normal audit test there is too much data to analyse and therefore we cannot see everything

Correct Incorrect Correct answer

Q10. What is the difference between the fields ZLAND and ZBNKS?

ZLAND is the country of the beneficiary and ZBNKS is the country of the bank

Correct Incorrect Correct answer

ZLAND is the country of the bank and ZBNKS is the country of the beneficiary

Correct Incorrect Correct answer

Q11. How can we add the user name to the REGUH table?

Through a join on the fiscal year, company code and payment document number

Correct Incorrect Correct answer

Through a join on the fiscal year

Correct Incorrect Correct answer

Q12. Which question does UMAP help us to answer?

How can we see records that have a similar set of values to a pre-identified exception?

Correct Incorrect Correct answer

Why did the record come out as having a rare score in the Isolation Forest?

Correct Incorrect Correct answer

WC094 – Did you know that you could use a simple python program to flag unusual payment transactions?

Participants 13

Webclass 94: Did you know that you could use a simple python program to flag unusual payment transactions?

admin June 9, 2026

Quiz Summary

Information

Results

Results

Categories

1. Question

2. Question

3. Question

4. Question

5. Question

6. Question

7. Question

8. Question

9. Question

10. Question

11. Question

12. Question