- Introduction
- Balance
- Clusters
- Concept drift
- Coverage
- Datasets
- General fields (previously entities)
- Labels (predictions, confidence levels, hierarchy, etc.)
- 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
- Create or delete a data source in the GUI
- Uploading a CSV file into a source
- Preparing data for .CSV upload
- Create a new dataset
- Multilingual sources and datasets
- Enabling sentiment on a dataset
- Amend dataset settings
- Delete messages via the UI
- Delete a dataset
- Export a dataset
- Using Exchange Integrations
- Model training and maintenance
- Understanding labels, general fields, and metadata
- Label hierarchy and best practices
- Analytics vs. automation use cases
- Turning your objectives into labels
- Overview of the model training process
- Generative Annotation (NEW)
- Dastaset status
- Model training and annotating best practice
- Training with label sentiment analysis enabled
- Understanding data requirements
- 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
Communications Mining User Guide
Defining taxonomy
To ensure the model returns the most relevant predictions, you must define your taxonomy correctly for each dataset. This is a critical step in the generative extraction process, because the learning and predictions of the model are based on the details you define beforehand.
- Log in to Communications Mining to land on the datasets page.
- Select New dataset to create a new dataset, and fill in the required fields. For more details, check Create a new dataset.
- Select Next.
- In the Define label step, select Add label.
- Define and structure the label
correctly, and provide a clear description of what it should consist of.
For example, you could add individual labels, without a structure, such as Label 1, Label 2, or you could create a hierarchy with the structure
Parent label > Branch label > Child label. This label hierarchy represents the taxonomy for this particular dataset. For more details, check Label hierarchy. - Select Create.
Once you set up your taxonomy, determine the processes you want to automate.
Next, identify the specific data points or fields that must be extracted from the documents.
Finally, organize the data points into a coherent structure (i.e. extraction schema), which will guide the data extraction process. To build the extraction schema, set up the fields through the Train, Explore, or Settings tabs, and set up the field types.
For more details, check the following resources:
