# Process Mining

## Process mining specific settings <a href="#lfd5me13jmt5" id="lfd5me13jmt5"></a>

This chapter introduces process mining-specific tagging conventions that allow you to customize the process mining analysis in the setting file.

#### Adding extra values to process step name <a href="#id-99v2812ptbkc" id="id-99v2812ptbkc"></a>

To add extra values to a step name without using a content-category tag, you can add process\_step\_name-property for the tag:

```json
{
"process_step_name": "{process_identifier_name}"
"key": "...",
"value": "..."
}
```

## Tagging a window as a separate process step <a href="#n269857mu2z3" id="n269857mu2z3"></a>

You can tag windows without process identifiers to be separate process steps. If these windows are found from multisystem workflows or between a start window and a process step, they get marked as individual process steps. These windows get the same transaction id as the process step that follows them.

```json
{
"key": "{process_identifier_name}",
"value": "PROCESS_STEP"
}
```

## A start window for a process sequence <a href="#id-4adsi6cj6iv" id="id-4adsi6cj6iv"></a>

Processing a transaction can start from a window appearing before detecting the identifier from the main process system. In these cases, you can mark processing beginning from a specific window and ending in a window where the identifier is present. The main use case is if extra steps exist between the start window and the main process window. The start window notation allows you to capture the whole task between the start window and the main process window as separate process steps.

You can mark the processing starting window by adding a tag:

```json
{
"key": "{process_identifier_name}",
"value": "STEP_START_SEPARATOR"
}
```

{% hint style="info" %}
*Note*: Separator steps are not marked as process steps by default. If you want to use separator steps as process steps, also add process step tag to the tagging rule.
{% endhint %}

#### Tagging a window as a shared preparation step for multiple transactions <a href="#s1gyuuow87u7" id="s1gyuuow87u7"></a>

If multiple transactions are processed in the same session and have the same preparation step, this preparation step can be tagged as follows:

```json
{
"key": "{process_identifier_name}",
"value": "PREPARATION_STEP"
}
```

## Adding a state tag or identifier for the event <a href="#ail5wfvm03y1" id="ail5wfvm03y1"></a>

1. Add a state tag key in the Dashboard for a process where you want to track a state.
2. Add tag or Extracted Identifier with the matching key/identifier\_name:

Example:

1. Add salesforce\_case\_state as the state name for a process in the Dashboard.
2. Add Tag or Extracted Identifier:

Tag:

```json
{
"key": "salesforce_case_state",
"value": "pending"
}
```

Extracted Identifier:

```json
{
identifier_name: "salesforce_case_state"
value: "pending"
}
```

**The result**: A step associated with the above tag/identifier is marked with the state pending.

## Configure Dashboard: process mining settings <a href="#asap5zdfa0oc" id="asap5zdfa0oc"></a>

Finally, configure the process mining features on Dashboard. Specifically:

1. Create a list of identifiers found in the data: You can either check the identifiers in the company’s settings, or query MongoDB for unique identifiers.
2. Review what identifiers should be added as processes in the Dashboard: Not all identifiers make sense as case identifiers for process mining. Review the list of identifiers with the business to ensure the correct process visualizations will be created.
3. Add processes to the Dashboard: Configure the process mining settings on the Dashboard to match the customer's requirements. This includes mapping extracted identifier names to process names to be displayed on the Dashboard.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://processmaker.gitbook.io/process-intelligence/developer-documentation/data-collection-and-configuration/process-mining.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.