Communications Mining user guide

Generative Annotation features are automatically enabled on datasets, no further action required.

Once a dataset is created, cluster suggestions are automatically generated within a short period of time. If a taxonomy has been uploaded, which is highly recommended, Communications Mining™ suggests both existing and new labels for clusters.

When you upload a taxonomy to a dataset, this also automatically triggers an initial model to be trained with no training data, only using label names and descriptions. This action may take a few minutes from when you have uploaded the taxonomy.

Cluster suggestions will appear for each Cluster page. This can be one or multiple suggested labels for each cluster.

If you have Label sentiment analysis enabled, Cluster Suggestions will have a positive or negative sentiment, which can be highlighted in green or red.

To identify an AI-suggested label, check the following image:

Model trainers should review each cluster suggestion, and perform one of the following::

Cluster suggestions can significantly speed up the first phase of the model training process by automatically generating suggested labels for each cluster. It can also help with taxonomy design, if users are struggling to define the concepts they want to train.

Cluster suggestions are generated based on the identified theme shared across the messages within a cluster.

The creation of clusters and generation of label suggestions is an automatic and completely unsupervised process with no human input required.

Label suggestions on clusters will be generated with or without a pre-defined taxonomy, but suggestions will be influenced and typically made more helpful by leveraging imported or existing labels.

Once the initial model has automatically trained using label names and descriptions as it's training input, predictions will appear for many of the messages in the dataset.

These predictions work in the exact same way as they have done previously, meaning that they are just generated with no training data.

If you have Label sentiment analysis enabled, initial predictions will have either a positive or negative sentiment in different shades of green or red, depending the confidence level.

Assisted Annotating works in any training batch or mode but it is most effective to use in Shuffle and Teach Label. You should follow the regular annotating steps in each training batch in the Train or Explore tabs.

Assisted Annotating can significantly speed up the second phase of the model training process by automatically generating predictions for each label with sufficient context, with no training examples required.

Initial predictions will be driven by the quality of the label names and natural language descriptions, such as vague names might lead to vague or minimal predictions. Detailed label descriptions can boost the initial performance of the model.

As you train your dataset further, the platform uses both the label names and descriptions and your pinned examples to generate relevant label predictions.

These will keep improving with more training and ultimately rely only on annotated training examples when enough have been provided.

Assisted Annotating still requires supervised learning by accepting or rejecting the predictions, but it accelerates the most time-consuming part of model training by providing better predictions with zero or very few pinned examples.