- Introduction
- Setting up your account
- Balance
- Clusters
- Concept drift
- Coverage
- Datasets
- General fields
- Labels (predictions, confidence levels, label hierarchy, and label sentiment)
- Models
- Streams
- Model Rating
- Projects
- Precision
- Recall
- Annotated and unannotated messages
- Extraction Fields
- Sources
- Taxonomies
- Training
- True and false positive and negative predictions
- Validation
- Messages
- Access Control and Administration
- Manage sources and datasets
- Understanding the data structure and permissions
- Creating or deleting a data source in the GUI
- Uploading a CSV file into a source
- Preparing data for .CSV upload
- Creating a dataset
- Multilingual sources and datasets
- Enabling sentiment on a dataset
- Amending dataset settings
- Deleting a message
- Deleting a dataset
- Exporting a dataset
- Using Exchange integrations
- Model training and maintenance
- Understanding labels, general fields, and metadata
- Label hierarchy and best practices
- Comparing analytics and automation use cases
- Turning your objectives into labels
- Overview of the model training process
- Generative Annotation
- Dastaset status
- Model training and annotating best practice
- Training with label sentiment analysis enabled
- Training chat and calls data
- Understanding data requirements
- Train
- Introduction to Refine
- Precision and recall explained
- Precision and Recall
- How validation works
- Understanding and improving model performance
- Reasons for label 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™
- Developer
- Exchange Integration with Azure service user
- Exchange Integration with Azure Application Authentication
- Exchange Integration with Azure Application Authentication and Graph
- Fetching data for Tableau with Python
- Elasticsearch integration
- Self-hosted Exchange integration
- UiPath® Automation Framework
- UiPath® Marketplace activities
- UiPath® official activities
- How machines learn to understand words: a guide to embeddings in NLP
- Prompt-based learning with Transformers
- Efficient Transformers II: knowledge distillation & fine-tuning
- Efficient Transformers I: attention mechanisms
- Deep hierarchical unsupervised intent modelling: getting value without training data
- Fixing annotating bias with Communications Mining™
- Active learning: better ML models in less time
- It's all in the numbers - assessing model performance with metrics
- Why model validation is important
- Comparing Communications Mining™ and Google AutoML for conversational data intelligence
- Licensing
- FAQs and more

Communications Mining user guide
Analytics FAQs
Dashboards
Dashboards are specific to the dataset itself. You can think of the Dashboard page as a place to create fully customisable reports or even a homepage for your datasets. Here you can add high-level information on the datasets, such as accuracy of the model, various charts, recent messages and the health of your taxonomy, etc.
For more information, check Reports.
Reports
Reports are message and thread-level reports, if the data is in thread form (e.g. call transcripts and email chains). If not, the message filter will be the default.
For more information, check Reports.
Dashboards are designed to be super flexible and fully customisable.
To create a card on your dashboard, create the filtered card on the Reports page and select the Add to dashboard button in the top hand corner. If you want to make a change to the filters of a specific card, you can do so directly from the dashboard.
You can filter specific charts further within the dashboard, and also have access to a global filters feature, which allows you to augment any existing filters across the existing cards on the dashboard, by narrowing down to a specific time-range, for example.
For more details, check Using dashboards.
The Compare page can be used to compare cohorts of messages or threads and conduct A/B testing. This functionality has many useful applications, such as testing the impact of a new marketing initiative or change programme over a time series, or between different regions or customer groups.
You can then filter the two different cohorts, A and B, using the filter bar.
For more details, check Using Comparison.
Filters narrow down the number of messages (i.e. messages) or threads (i.e. conversations) that are in scope for your analysis or exploration, allowing you to target very specific segments or cohorts of your dataset.
Charting specific labels or values just determines which labels or values are shown on the charts, it doesn't limit the scope (i.e. number of messages or threads) that the platform is considering when rendering the charts in Reports.
A quick summary
Chart labels - Used to select which labels actually appear or do not appear in the charts in the Reports page.
Chart values - Used to chart specific string property values on a chart in the Segments section of the Reports page. For more details, check Charting specific labels and property values.
Label filter - Works the same way as in Explore, where you can use the label filter bar to filter for messages that have or do not have specific labels predicted.
Property Filter - Works the same way as in Explore, where you can use the property filter to filter by specific properties which includes metadata, dates, and so on. For more details, check Applying filters in Reports.
For more details on how to apply filters generally, check Applying filters in Reports.
For more details on how to apply advanced prediction filters, that is, multiple label filters, check Advanced prediction filters.
Messages allows you to apply analytics at the level of conversations, in addition to 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.
For more details, check Applying filters in Reports.
You can apply as many filters as you like in the filter bar to get as specific as you need to with your analysis.
As a rule of thumb, different filter types get combined, that is, they create AND statements, together if multiple filters are applied, narrowing down the selection each time another filter type is added.
Different values within a filter type, except for labels, get grouped together, that is, they create an OR statement. For example, if you apply a Sender Domain filter for emails, and select multiple values, the platform will filter to messages or threads that have any of the selected values as the sender domain.
Labels are the exception as you can apply AND as well as OR statements with label filters, by adding multiple ones. For more details, check Advanced prediction filters.
If you do not have a time series filter applied to your charts, every chart will automatically update as new data is added. This is dependent on having a live integration for your data into the platform.
You can also add dynamic time filters to charts in your dashboard, such as last 7, 30, 90 or 365 days. These will also automatically update as time passes and newer data is added.
To export data from individual reports, select the three dots in the top right of a given report and select Download Chart as CSV. The CSV generated will generate outputs based on the report that is visualised.
The platform makes all of the data in a dataset accessible via the API, which can be integrated to downstream databases or directly to visualisation tools. To find out more about integrating Communications Mining™, check the Communications Mining developer guide.
+
button in the Dashboard
page. You will be presented with the option to Clone this dashboard or
Create from scratch.
No, you can add as many charts as you like to your dashboard, and you can create as many dashboards as you like for each dataset.
Label Summary - Shows charts of label volumes.
Trends - Shows charts of message volumes, by label.
Segments - Shows charts of label volumes split by message metadata category, also known as user properties.
Threads - Shows charts of thread volumes.
Thread charts and filters are only available when the data is aggregated by threads, that is, by conversations, rather than by messages, that is, by individual messages. Typical examples are email threads, chat conversations, and call transcripts.
- What's the difference between Dashboard and Reports? When should I use them?
- How do I manage my dashboards?
- What does Compare do? How do I use it?
- What is the difference between a label or property filter and Chart labels or Chart values?
- How do I apply filters?
- What does the message or thread filter do?
- How many filters can I apply in the filter bar?
- How do multiple filters combine with each other?
- How do I create Dashboard charts that update automatically based on the latest data?
- How do I export the data from Reports?
- How do I set up a live connection between the data outputs, and my own visualisation tool?
- Can I create more than one Dashboard?
- Is there a maximum number of reports that I can include in the Dashboard?
- In Reports, what is the difference between Label Summary, Trends, Segments, and Threads?
- Why is Threads not available to me under Reports?
- I have Threads enabled on my dataset. Why can't I access the Threads tab on the Reports page?