- Getting started
- Balance
- Clusters
- Concept drift
- Coverage
- Datasets
- General fields (previously entities)
- Labels (predictions, confidence levels, hierarchy, etc.)
- Models
- Streams
- Model Rating
- Projects
- Precision
- Recall
- Reviewed and unreviewed messages
- Sources
- Taxonomies
- Training
- True and false positive and negative predictions
- Validation
- Messages
- Administration
- Manage sources and datasets
- Understanding the data structure and permissions
- Create a data source in the GUI
- Uploading a CSV file into a source
- Create a new dataset
- Multilingual sources and datasets
- Enabling sentiment on a dataset
- Amend a dataset's settings
- Delete messages via the UI
- Delete a dataset
- Delete a source
- Export a dataset
- Using Exchange Integrations
- Preparing data for .CSV upload
- Model training and maintenance
- Understanding labels, general fields and metadata
- Label hierarchy and best practice
- Defining your taxonomy objectives
- Analytics vs. automation use cases
- Turning your objectives into labels
- Building your taxonomy structure
- Taxonomy design best practice
- Importing your taxonomy
- Overview of the model training process
- Generative Annotation (NEW)
- Understanding the status of your dataset
- Model training and annotating best practice
- Training with label sentiment analysis enabled
- Train
- Introduction to Refine
- Precision and recall explained
- Precision and recall
- How does Validation work?
- Understanding and improving model performance
- Why might a label have low average precision?
- Training using Check label and Missed label
- Training using Teach label (Refine)
- Training using Search (Refine)
- Understanding and increasing coverage
- Improving Balance and using Rebalance
- When to stop training your model
- Using general fields
- Generative extraction
- Using analytics and monitoring
- Automations and Communications Mining
- Licensing information
- FAQs and more
Filtering Reports
In Reports you can apply filters to enhance data insights.
User permissions required: ‘View Sources’ AND ‘View Labels’.
In Reports you can use the filter bar to manipulate the charts and update the summary statistics presented.
The filter bar works mostly in the same way as in Explore. The key difference is that in Reports, you can use thread filters for conversational datasets containing threads (including email threads, chats and call transcripts). See below more details on thread filters.
It's important to remember that you can add multiple filters (including multiple metadata fields) in Reports (and Explore), and that all of the filters will determine the messages in scope (as shown by the summary stats at the top of the page).
When selecting which labels to appear in the label-specific charts, use the Charted labels dropdown menu in the top bar on either the Trends or Segments page.
The label filter bar in Reports works in the same way as in Explore. For more information and a detailed explanation of the Advanced Prediction filters check Advanced prediction filters.
The label filters do not determine which labels are actually plotted on the label-specific chart. This is determined by the 'Charted labels' selector. See below for more detail.
At the top of the filter bar, you can toggle between Messages and Threads. This allows you to apply analytics at the level of conversations, instead of individual messages.
The threads filter is applicable for longer-form conversations like email threads, phone calls, and live chats. Once the Threads filter is selected, a fourth reports tab will become available, containing charts relating to thread properties and label volumes.
The attachment based properties are available for each message that supports attachments.
These properties include:
- Has Attachment (Yes/No)
- Number of attachments
- Attachment Size (Mb)
- Total Attachment Size (Mb)
- Attachment Name Includes - Provides a fuzzy search capability
There are two methods for filtering the time series within Reports, using the time series filter in the filter bar, or selecting an area on a time series chart.
This will filter the messages in scope to only the select period, as well as adjust any time series charts to the filtered period.
Using the time series filter in the filter bar is the same as in Explore - simply select a 'From' and/or a 'To', or choose a preset custom range such as 'Last 7 days'.
To select the area on a time series chart:
- Select a chart that shows a trend over time
- Highlight a time period on the chart by clicking your mouse at the start of the time period you want to filter and drag it to the point in time you want to finish on
- This will filter the chart for the specific time period that you highlighted
Once the chart refreshes, you will see the time period that you selected has automatically been applied to the timestamp in the filter bar. You can see the time period in the summary statistics, at the top of the page.
In the Trends tab, you can select which time sequence you want the time series charts to be displayed in using the dropdown menu just below the summary statistics.
This allows you to plot the time series charts as:
- Daily
- Weekly
- Monthly
- Yearly
Advanced prediction filters represent a significant step forward in how the label filters work in both the Explore and Reports pages in the platform.
We've significantly expanded the platform's capability to support intelligent filter combinations on label predictions.
Here are some of the things you can now do when filtering by label predictions:
- Apply multiple label filters at once, in both Explore and Reports
- Filter for messages that have one of a number of selected labels predicted (i.e., Label X OR Label Y OR ...)
- Filter for messages that have multiple different labels predicted (i.e. Label X AND Label Y AND ...)
- Filter for messages that do not have certain label(s) predicted (i.e. NOT Label Y)
- Search for messages containing specific search terms, whilst having label filters applied
Essentially, the new filters let you apply any combination of 'AND', 'OR' and 'NOT' when applying more than one label filter. These filters can give you much greater flexibility when training and interpreting your data, and can provide much deeper insights on what's happening in your communication channels.
As a result of these changes, we've also updated the way you select a label to sort by when training in Explore, and how you select labels to appear in the charts in Reports.
When you filter to a label now in Explore, this no longer determines the training mode or the sort order of the current training mode. Instead this is determined by a taxonomy dropdown menu that appears when you select a training mode that is specific to a certain label (e.g. 'Check label' or 'Missed label').
This is to enable you to be sort by a certain label for a training mode and have an additional label filter applied.
How this works in practice is discussed in more detail further down this article.
Similarly, when you filter to labels in Reports this no longer determines which labels appear in charts. Instead, this also acts exclusively as a filter and the dropdown 'Charted labels' menu determine which labels are shown in the charts. How this works in practice is also discussed in more detail at the end of this article.
There are now two ways to apply label filters, and they can be used in combination with each other to create the right type of query.
Taxonomy Bar
The taxonomy of labels functions as a normal filter bar, and allows you to select multiple labels at once with a single click for each.
Selecting multiple labels from the taxonomy creates an 'OR' type query.
If you selected Label A, Label B and Label C in the taxonomy, this creates a 'Show me messages with Label A, Label B, or Label C predicted' query.
The following example would return messages that contained at least one of the three selected labels:
Add Label Filter
The second filter option is the + Add Label Filter button above the taxonomy bar.
This enables a dropdown taxonomy that allows you to select more complex filters, including excluding certain labels from consideration.
From this dropdown menu, you can select multiple labels that you want to include or exclude, by clicking the plus or minus buttons.
The result would be something like the following example, which would return messages predicted to have the label 'Billing/Refund', but that also do not have the label 'Chaser/Urgent' predicted:
You can click + Add Label Filter multiple times to add additional layers to your query. Two separate label filters create an 'AND' type query, whilst multiple labels selected in the same label filter create an 'OR' type query.
In the example below, multiple label filters have been applied individually. This creates a filter that will return messages predicted to have either of the 2 labels in the first filter, but that also have the 'Policy' label predicted, and do not have the 'Quote' label predicted.
A helpful tip is that by clicking the '&' sign in an individual filter containing multiple labels, you can automatically split them out into individual filters. This would change the query from 'OR' (i.e. any of these labels predicted) to 'AND' (i.e. all of these labels predicted).
Combining taxonomy bar filters and added label filters
It's possible to combine filters from both the taxonomy bar, and individually added label filters. Filters applied in the taxonomy bar are treated as an 'AND' query with any individually applied label filters.
Combining label filters and sorting by label for training
What these new filters also mean, is that you can now apply label filters and sort by a specific label for a training mode.
To enable the advanced prediction filters, a number of other updates have been made that do change the way you select labels to train in Explore. Overall, these changes make the platform more consistent across the platform, as well as providing greater capabilities to run detailed queries on your data.
The top bar in Explore now shows the selected training mode and sort order in the top left, with the search bar on the top right (see below).
Importantly, when selecting a specific label to train you no longer use the filter bar on the left, but rather you select the training mode first and a taxonomy dropdown then prompts you to select the label you wish to sort by.
The following example shows a user having selected 'Check Label' and then being prompted to select which label to sort by:
You can also see that the pinned label counts, red dial indicators, and any performance warnings for a label are also displayed in the taxonomy dropdown, giving you all the information you need to select the right label to sort by.
Once a label is selected, it will be shown next to the training mode alongside the number of pinned examples and any performance warning indicators.
Now, if you select a label from the label filter bar, the platform will prompt you to select it from the sort order dropdown menu if you intend to train the selected label:
You can see a specific mode in Explore, 'Sentiment', to sort by positive or negative sentiment only (see below). If you want to sort by label(s) and sentiment, you can still do that using the 'Label' sort order and adding a sentiment filter:
The label filters act exclusively as filters to update the messages that are in consideration (much like any other user property filter) by including or removing messages that have or don't have certain labels predicted.
To select which labels actually appear in the charts in the Reports page, there is now a 'Charted labels' dropdown menu in the top bar.
Simply select whichever combination of labels you want to appear and these will be plotted on the charts which distinguish between different labels, as in the example below.
Remember, this does not change the messages that are in consideration by the platform, merely which labels are plotted on the label specific charts. An easy way to remind yourself of this is to note that the total number of messages shown at the top of the page does not change when you select any labels from the 'Charted labels' dropdown.
All of the labels that you select will appear highlighted at the top of the page as a visual reminder, and can easily be removed by clicking the 'x' icon next to each label name.
User permissions required: ‘View Sources’ AND ‘View Labels’.
Near the top right of the page on the 'Label Summary', 'Trends', 'Segments', 'Threads', and 'Comparison' page is the 'Chart labels' dropdown option, which you can use to select which labels appear or do not appear in the charts in the Reports page
Simply select whichever combination of labels you want to appear (the plus symbol) and these will be plotted on the charts that distinguish between different labels, as in the example below. Alternatively, you can select labels you don't want to appear by using the exclude button (the minus symbol).
If you select a number of labels, they will appear in a bar at the top of the page (see below) as a visual indicator of which labels have been selected to appear on the charts.
All of the labels that you select or exclude will appear highlighted at the top of the page as a visual reminder, and can easily be removed by clicking the 'x' icon next to each label name.
This action does not change the messages that are in consideration by the platform, merely which labels are plotted on the label specific charts.
An easy way to remind yourself of this is to note that the total number of messages shown at the top of the page does not change when you select any labels from the 'Charted labels' dropdown.
In Segments, much like you can choose to chart labels (see above), you can also chart specific string property values.
You can click a number of property values you wish to plot on the chart, and this will override the default values plotted, which are typically the property values that have the highest corresponding message volumes.
In the example below, the email property of 'Sender Domain' is shown. Using the dropdown at the top-right of the chart, you can select a number values to plot instead of the default ones shown.
User permissions required: ‘View Sources’ AND ‘View Labels’.
Beneath each chart you’ll see a number of buttons when you hover over it with your mouse. These buttons allow you to change the chart view and are tailored to each graph.
The buttons available change depending on the exact chart type. They are a reflection of the current view shown in the chart, and so in some instances clicking them will change the mode to an alternative view, and the button displayed will also update.
Check out the table below to see the chart manipulation buttons.
| Add this chart to your dashboard | |
| Option to download an image of the chart, download the raw data for the chart as CSV, or add chart to your dashboard | |
| Show percentages | |
| Show totals | |
| Display as column/bar chart | |
| Display as line chart | |
| Display as scatter graph | |
| Display as doughnut chart | |
| Sort by prominence | |
| Sort alphabetically | |
| Sort reverse alphabetically | |
| View labels with the biggest changes in rank | |
| Change the number of labels displayed |
- Applying filters in Reports
- The filter bar and adding or removing filters
- Label filter
- Threads filter
- Attachment filter
- Time series filtering
- Chart time sequence
- Advanced prediction filters
- Overview
- Impact on model training and filtering charts
- Applying advanced prediction filters
- Sorting by label for training modes in Explore
- Selecting labels to appear in charts in Reports
- Charting specific labels and property values
- Charting specific labels
- Chart specific property values
- Chart buttons
