> ## Documentation Index
> Fetch the complete documentation index at: https://koreai-ai-for-process-dev.mintlify.app/llms.txt
> Use this file to discover all available pages before exploring further.
# Conversation Mining - Interactions
The Interactions page lets supervisors review and filter scored interactions within their assigned queues. It provides insights into conversation quality, agent performance, and customer experience — helping identify which interactions need coaching or follow-up.
Navigate to **Quality AI** > **ANALYZE** > **Conversation Mining** > **Interactions**.
***
## Key Capabilities
* **Review and filter interactions**: Access scored interactions in assigned queues to find conversations needing coaching or attention.
* **Monitor conversation quality**: Get a clear view of agent performance and customer experience at the conversation and agent level.
* **Gain deeper insights**: Use taxonomy-based Topic Discovery, topic-level sentiment, and resolution insights.
* **Filter by experience insights**: Apply filters for configured intents (L1, L2, L3), generated intents, and overall resolution status.
* **Audit with precision**: Customize the view using metadata and column controls, including configured intent and resolution columns.
* **Refine searches**: Use keyword filters with speaker-specific options, include/exclude terms, and options to hide resolved cases.
* **Navigate by timeline**: Use timeline-based search and jump-to-match for fast, focused review.
* **Save custom views**: Store preferred filter sets and column configurations for repeatable audits.
***
## Interactions Dashboard
### Columns
| Column | Description |
| -------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Agents** | Name of the agent who last handled and ended the interaction. |
| **Processing Status** | Green = all processes succeeded; Red = at least one process failed. |
| **Actions** | Assign the interaction to a bookmark for future reference. |
| **Queues** | The queue where the interaction was terminated. The evaluation form used matches this queue. |
| **Kore Evaluation Score** | Auto QA score based on the completed evaluation form. Hover over the warning icon to see the API processing status. |
| **Sentiment Score** | System-generated score based on the customer's words. |
| **Supervisor Audit Score** | Score assigned by the supervisor for audited interactions. Manual Evaluation metrics display a distinct indicator. |
| **Start Time** | Conversation start time (for example, 24 May 2024, 1:17:10 PM). |
| **Duration** | Call duration including talk time, hold time, and ACW (for example, 0h 6m 25s). |
| **Moments** | Counts for adherence, violations, and omissions. Hover to see details: Questions Adherence, Violations, Omissions. Select an interaction to view Moments in the AI-Assisted Manual Audit page. [Learn more](../ai-assisted-manual-audit.md). |
| **Sentiment Trend** | Distribution of positive, negative, and neutral tones; tracks sentiment changes throughout the call. |
| **Emotions** | Customer emotional states detected during the conversation. |
| **Conversation ID** | User-defined identifier for custom tracking. |
| **Custom Conversation ID** | Unique system-generated identifier used as the primary key. |
| **Configured Intents** | Topics or intents across all levels (L1, L2, L3), color-coded by sentiment (red, green, gray). |
| **Generated Intents** | AI-identified intents for each conversation. |
You must create bookmarks in [Settings](../configure/quality-ai-general-settings.md) before tagging interactions for future reference.
### Dashboard Filters
| Filter | Description |
| -------------- | --------------------------------------------------------------------------------- |
| **Columns** | Include or exclude specific fields from the view. |
| **Bookmarks** | Assign interactions to bookmarks and view all bookmarks linked to an interaction. |
| **Date Range** | Select the date range for interactions; defaults to the last 7 days. |
If you attempt to evaluate an interaction not assigned to you, you cannot submit the evaluation.
***
## Filters
Use filters to narrow results to conversations that need attention. Add filters from the **Filters** dropdown and refine using multiple category panels.
### Add a New Filter
1. Select the **Filters** dropdown in the upper-right corner.
2. Select **+ Add New Filter**.
3. Choose a **Filter By** category.
***
## Filter Categories
### Filter by Efficiency
Filters interactions based on handling metrics and process adherence.
1. Select **Channels** (Chat or Voice) to limit to a specific communication mode.
2. Select **Audit Status** to view Audited, Assigned, or Not Assigned conversations.
3. Enter the **Conversation ID** or **Custom Conversation ID** to filter a specific interaction.
4. From **Queues**, select the routing group.
5. From **Agent Groups**, select the group linked to the chosen queue.
6. From **Agents**, select the agent assigned to the queue.
7. Enable additional options as needed:
* Filter by handling time
* Filter by deviation from AHT
* Deviation (%)
* Number of Transfers
### Filter by Experience
Filters based on customer experience metrics.
| Filter | Description |
| ---------------------- | ----------------------------------------------------------------------------------------------------------------------------------- |
| **Avg. Waiting Time** | Select a date range in seconds. |
| **Sentiment Score** | Slider to set min-max range for positive (higher) or negative (lower) sentiment. |
| **CSAT** | Distribution of interactions by customer satisfaction score. |
| **Configured Intents** | Filter by topic/intent across L1, L2, L3. Child lists update based on parent selection. |
| **Generated Intents** | Select configured or AI-suggested intents. |
| **Overall Resolution** | Filter by Resolved or Unresolved (both selected by default). A conversation is resolved when all applicable L3 intents are defined. |
| **Churn Monitor** | Filter by Churn Risk (level of customer churn risk) or Escalation (number of escalations raised). |
Select Configured Intent levels in order: L1 first, then L2, then L3. You must select at least one level to apply the filter.
### Filter by Behavior
Filters based on agent conduct and interaction dynamics.
| Filter | Description |
| ---------------------------- | --------------------------------------------------------------------------------------------------------------------------- |
| **Metric Name** | Filter by specific evaluation metrics using Pass/Fail radio buttons. |
| **Metric Qualification** | The selected metric appears as a tag; clear it with **X**. When opened from Adherence Heatmap, filters apply automatically. |
| **Language** | Select languages from the dropdown. |
| **Empathy Score** | Measures how well the agent responds to the customer's emotional state. Higher = more empathetic. |
| **Crutch Word Score** | Frequency of filler words used by the agent. Higher = more crutch words. |
| **Agent Playbook Adherence** | Adherence percentage to Agent AI playbooks for the interaction. |
| **Kore Evaluation Score** | Automated QA score based on the evaluation form assigned to the queue. |
### Filter by Keywords
Search, filter, and review conversations by specific keywords or phrases.
| Category | Details |
| ----------------------------------- | --------------------------------------------------------------------------------------------------------------------------- |
| **Select Conditions (AND/OR)** | **AND**: All specified keywords must appear. **OR**: At least one keyword must appear. |
| **Conversation Including Keywords** | Search by keyword or exact phrase with optional speaker filters (Agent only, Customer only, Both). Add up to 8 keywords. |
| **Conversation Excluding Keywords** | Enter keywords that must not appear. Use speaker filters. Add up to 5 exclusion keywords. All exclusions use AND NOT logic. |
| **Audit Navigation** | Supports transcript autoscroll and match relevance filtering. |
| **Apply Keywords** | Select **Apply Keywords** to activate filters, then **Apply** to run the search. |
| **Filtered Interactions** | Displays the count of interactions matching the selected keywords. |
***
## Manage Saved Filters
| Action | Description |
| ------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| **Apply Filters** | Stores current settings as an Unsaved Filter. Use **Save & Apply** to save and name filters for audit allocation. |
| **View Saved Filters** | Shows all saved filters and their associated unsaved filters. |
| **Filtered Interactions** | Real-time count of matching conversations. |
| **Customization Options** | Copy, Mark as Default, Edit, and Delete filters. |
If interactions are not used for audit allocation, you can apply filters without saving.
***
Built with [Mintlify](https://mintlify.com).