- 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
Best practices and considerations
The highlighted general field should cover the entire word, or several, in question, not just part of it. Do not include additional spaces at the end of the field.
Similar to labels, do not partially review your general and extraction fields.
- General fields are reviewed at the
paragraph level, not the entire message level. When you review a paragraph for
fields, review all the fields in the paragraph.
Not confirming a field in a paragraph where you have labelled other fields, tells the model that you do not consider it a genuine example of the predicted field. This is reflected in the validation scores and the general field performance.
- Extraction fields are reviewed at the
message level, not just the paragraph level. When you review an entire message
for fields, review all the fields in the message.
Not confirming a field in a message where you have labelled other fields, tells the model that you do not consider it a genuine example of the predicted field. This is reflected in the validation scores and extraction field performance.
- Global fields cannot overlap with each other, or with another example of itself.
- Global fields and Extraction fields can overlap with each other.
- You can use the same span of text as many times as needed by different extraction fields.
- There is currently no general field normalization preview in Communications Mining™. Fields that should be normalized will get normalized in the downstream response. Normalization in Communications Mining will be available in the model in the future.
- If a child label has extractions on it, its parent does not inherit the extraction examples automatically. For labels, its parent automatically inherits the extraction examples.
- Providing additional extraction examples does not improve the performance of a label. To improve the performance of a label, focus on label-specific training.
- Improving label performance allows you to
increase the likelihood that you capture occurrences where a label, and
subsequently its extractions, should have been predicted.
To improve the performance of your extractions, provide validated examples on the extractions itself.
