Introduction
Accounting professionals know that journal entries can be repetitive and prone to human error. Mistakes like incorrect account entries or swapped amounts can lead to discrepancies in financial records. The following tool is a beta version, and can be potentially used by small enterprises that do not have sophisticated ERPs. One can load the general ledger in CSV format. The code then runs and analyzes the past journal entries in general ledger to understand common patterns and flags entries that deviate from these norms.
How Does It Work?
It uses an algorithm known as the Isolation Forest. Think of it as a smart assistant that learns what your typical journal entries look like. It takes into account these following factors: dates, account names, debit amounts, credit amounts, and transaction descriptions. It then flags journals which appear unusual.
Tool Code
The beta (untested) code is included here. The input is a general ledger data in the CSV format, that contain the following headers in the exact order – Date, Account, Debit, Credit, Description. The input file looks as follows:
The sample general ledger CSV input file can be found here:
As for the Python code for the tool itself, it is included in the following file:
Based on the input file, the code then outputs the following transactions as unusual:
- Date: 2023-02-08
- Account: Cash
- Debit: 0
- Credit: 10500
- Description: Refund Processed
This entry differs from the normal pattern, not only in the ‘Debit’ and ‘Credit’ amounts but also in the ‘Description’.
Parameters
Certain parameters may need to be adjusted in the model. For example, contamination parameter in the line “model = IsolationForest(contamination=0.01, random_state=42)” is currently set to 1% which may not be optimal for all datasets and could result in false alarms.
Conclusion
The post provided a beta version (untested) of the tool for scanning general ledgers based on the dimensions Date, Account, Debit, Credit, Description. Certain calibration may be required for different GL data sets.