- 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
This section describes the relationship between key platform components such as integrations, mailboxes, buckets, sources, and datasets to help you set up your data effectively:
- Set up an Exchange integration through a service account, which syncs data from
the Microsoft Exchange Server.
This integration allows you to sync multiple mailboxes, which are each stored in a bucket, and each bucket can contain multiple mailboxes.
- Set up a source, which is a collection of raw annotated communications data of a
similar type.
When setting up a source, if you are using data from an email integration, you must specify which bucket you want to sync from, that is the bucket where the mailboxes in scope for your use case are stored.
- Once you have finished setting up your source, add your source to a dataset,
which is where your Model will be trained.
Each dataset belongs to a project, which is a permissioned storage area within the platform. Each dataset and source belongs to a specific project, which is designated when they are created.
The following diagram illustrates how all these components are related:
Apply the the following steps in this specific order so that data from your mailboxes shows up in the platform.
Create a new integration
An integration refers to the connection and sync of data from the Microsoft Exchange service. You can set up a single integration for an Exchange service account, which can contain multiple mailboxes within it.
To create the integration, proceed as follows:
- Navigate to the Administration page.
- Select the Integrations tab, which allows you to set up a live connection
with Microsoft Exchange. For more details on how to set up your integrations,
check Integrations overview.
Add and configure one or more mailboxes
Once you have set up the integration with Microsoft Exchange, you can add the mailboxes in scope for your use case:
- Select the email icon on your integration card.
- Select Add mailbox, and fill in the mailbox details:
- Mailbox name - Enter the Email want to add to the integration.
- Bucket - A bucket is the location where one or more of your
mailboxes will be stored, and each bucket can contain multiple
mailboxes. Select one of the following options:
- New bucket:
- Create new bucket Mailbox
- User define new bucket - Select the Project, enter a Name, and, optionally, a Title.
- Existing bucket - Select an existing bucket from the dropdown list.
Make sure you select the appropriate bucket because:
- You must always add mailboxes from the same bucket into the same dataset.
- If you
need to sync a large number of mailboxes and plan to include
them in the same dataset, the best practice is to place them in
the same bucket.
Note: If you need to sync a large number of mailboxes to a bucket, and have over 2000 mailboxes, you are recommended to contact your UiPath® account manager. This allows Communications Mining™ to give you the best performance.
- New bucket:
- Time filters:
- From timestamp - Enter a timestamp in the Sync from timestamp field to specify when email syncing should begin.
- All time -
This option will sync all available emails, regardless of when
they were received.
Important: Selecting the All time option may lead to syncing more data than expected, and messages are charged on upload.Note: If you do not apply any filters, all emails will be synced, including deleted items.
- Folder filters - Enter the Allowed folders or Denied folders.
- Participant filters - Enter the Allowed participants or Denied participants.
- Attachments -
Enable or disable the Sync attachment contents option to sync and
store attachments. You can edit this option for new emails at any time.
Existing emails are not affected.
Note: Regardless of the selection, Communications Mining™ will always capture attachment metadata.
- Select Add mailbox to complete the process.
- If you change your mind, and you no longer want to add the new mailbox, select Discard new mailbox.
Repeat the steps for each mailbox that you need to sync.
Create a new source, specifying the Bucket where your mailbox is located
Create a new source in the platform as described in the Creating or deleting a data source page. Each data source will only have one bucket associated with it, at a maximum.
Create a dataset, and add your source
Create your dataset, and add to it the source containing the bucket where your mailboxes are synced. For more details, check Creating a new dataset.
If you have an existing dataset to which you want to add the source containing the integration, modify the dataset settings, and add the new source.
