Has no need for seed sets
Next generation Continuous Active Learning Predictive Analytics
Uses advanced modeling techniques to build and apply models to quickly find responsive documents for all tags
Simultaneously trains multiple models (a model for each tag)
Uses a proprietary workflow with unique capabilities that lets users train a software model by reviewing and tagging a small portion of the data set, and then use the trained model to automatically generate predictive results for the remaining documents
Allows having only a few reviewers regardless of the size of a document set
Performs real-time monitoring of model building
Saves models for use in subsequent productions or matters/cases
SentioAI uses machine learning to find relevant documents based on input provided by users. Analogous to a music streaming service "choosing" which songs a user will enjoy based on previous song selections, SentioAI uses document tags to train a ranking algorithm that orders relevant documents from most to least likely.
In addition to being able to train models using only a small data set, SentioAI's proprietary workflow enables having only a few reviewers regardless of the size of a document set. The software also detects documents that are likely misclassified by a reviewer and recommends verification.
The software provides users with the ability to algorithmically eliminate human review of a large percentage of a document collection - often by as much as 90%, during the first-pass review, potentially reducing a company's document review and management expense by millions of dollars and achieving improved results in far less time.
If a seed has already been created SentioAI can also be used to run TAR.
SentioAI works both as a stand-alone product and is integrated with Relativity®. The software can run in the cloud or behind a client's firewall.