The article talks about the intelligent sampler tool.
What's the Intelligent Sampler?
MindBridge’s Intelligent Sampler is a tool used to generate samples in the Data Table. You can use the Intelligent Sampler to create samples that are either:
- Risk-Stratified: This method uses the risk score of entries to stratify the population generated by the applied filter. This method is unique to MindBridge and favors high risk and medium risk entries when creating the sample.
- Random: This method creates a random sample from the population generated by the applied filter. Unlike the risk-rated method, each item in the population has an equal chance of being selected, regardless of their risk level.
The Intelligent Sampler’s 60/40 Split
When the Intelligent Sampler generates a sample, it selects all of the high risk entries available to be sampled, up to the selected sample size; if there are not enough high risk entries to fill the sample, the rest of the sample is split to contain 60% medium risk entries and 40% low risk.
For example, if 100 entries were selected for the sample size and the data contains 21 high risk entries, all 21 high risk entries would be added to the sample. Of the remaining 79 possible entries left for this sample, 47 (60%) are medium risk entries while 32 (40%) are low risk entries. These 47 medium risk entries are randomly selected from the remaining 841, as with the 32 entries from the 32,179 low risk entries.
How does the Intelligent Sampler Randomly Select Items?
When the Intelligent Sampler is launched, the data randomization process begins. Our method for randomizing the data is as follows:
The selected population is loaded from the dataset, and a Pseudorandom Number Generator (PRNG) is used to sort the population.
- The PRNG is configured with a random seed based on system time and other variables;
- A random number is generated for each item using the PRNG;
- The items are sorted by their generated numbers; and
- The lowest "n" (sample size) items are returned as the sample.
For risk-stratified sampling, the above process runs three times, for high, medium, and low-risk entries. For random sampling it runs once across the whole population.
After sampling, the Intelligent Sampler will display the sampled total debits, credits, net value, and absolute value.
Clicking Add to Audit Plan will add each entry to the audit plan with an “Open” status.
Sampling Outstanding Entries in Accounts Payable
When the intelligent sampler is selected from a data table of outstanding entries, samples are made from your client’s end-of-year open payables list.
The intelligent sampler works the same on outstanding entries as it does for GL or AP detail entries, with one exception. Some outstanding entries are not risk scored, because these entries do not match any single entry in the AP detail, and these un-scored entries are omitted from the intelligent sampler. The random sampler treats scored and un-scored entries equally when choosing a sample.