Viewing Predictive Scores as a Metadata Field
When a case has Artificial Intelligence enabled, scores and tags will be synced between the Review tool and the analytic engine at a near real-time pace. When an AI model is created, there is a corresponding field created by default in the Review tool. This field will be named with the Tag or Choice name, prefixed by AI Score - <Tag Choice> or Nexlp <Fieldname>. This field can be added to a field profile for use in searching and filtering.
Artificial intelligence enabled tag metadata fields:
Artificial intelligence analytic engine linked field metadata:
Tag AI Predictive Scores
When the analytic engine has a score to associate with a document, it will be stored in this integer field. The values can range from 0-100. These scores will be shown as parentheticals following a Mutually Exclusive Tag or a Multi-Select or Tree Tag Choice, and as integer values in the related AI Score fields.
Here are predictive AI scores for a document shown in the Tag pane:
Here are some of the comparable scores in metadata for this document - note the scores match for each labeled item:
Note
Sometimes predictive AI scores will be negative numbers, which reflect the below error codes.
-100 - Unclassified: These are the documents that get a probability score of 50 from the COSMIC classifier. This means they are classified as neither positive nor negative as per the current round of COSMIC classification.
-200 - No Score – Errored Documents: An error has occurred during the 2nd pass of the processing for these documents.
-300 - No Score – Empty Documents: When a document is found to have no text and metadata representation in the form of vectors. This almost never happens for emails as there is metadata in the form of: From, Sent, Subject, etc. But this can happen for an attachment that has a unique filename.
Tip
In order to reduce this number, tag more documents and train the model. As the model grows stronger it will reduce the number of empty documents by being able to recognize the metadata and text features in order to give the documents a proper score.
-400 - No Score – Missing Text Vectors: This number reflects the number of documents for which our system cannot locate text vectors. This can occur if something abnormal happened during the processing or if any vector files have changed location at any point.
-500 - Uncertain – No Model Features: When a document does have some metadata or text features but doesn’t have those features in the current run of the COSMIC classification model. In other words, if the model doesn’t have the set of features that a document has, we cannot score the document against the model so it gets marked as “Empty”.
Analytic Engine Linked Custom Field Metadata
Where additional analysis has been undertaken on the AI side and written to custom fields, those fields will be replicated in Reveal Review with a NexLP prefix to carry the data across the link.
Users with Prediction AI access permission (set in Project Admin > Tags) and others with AI access rights may log into the related database in the Artificial Intelligence module and further examine the analytical information.
To examine the Cluster referenced here, for example:
Open Artificial Intelligence under the Flyout Menu.
Login with your AI username and password.
Select the Storybook with the same name as your Reveal Review project.
To examine Clusters, open Entities on the dropdown menu and enter either the number of one of the keywords indicated in the Nexlp Cluster field.
More information on this topic will be found in the Reveal AI documentation.