In his recent post, A TAR is Born: Continuous Active Learning Brings Increased Savings While Solving Real-World Review Problems, Catalyst founder John Tredennick describes Continuous Active Learning (CAL) as a proposed advanced protocol for Technology Assisted Review (TAR). Developed by Maura Grossman and Gordon Cormack, CAL integrates training and review in a process where the platform’s understanding of the subject matter continuously evolves as documents are reviewed and as new documents are ingested. Tredennick compares CAL to Simple Passive Learning and Simple Active Learning protocols and outlines how CAL evolves TAR with further cost and time savings, improved results and elimination on the reliance on SME’s.
It is possible to see the CAL method as a common-sense progression and evolution of TAR. CAL more closely resembles the holistic, iterative process of an efficient, traditional document review where case development and understanding gathers steam as the review “machine” learns as it reviews. Furthermore, CAL, as a hybrid approach, may alleviate some of the remaining resistance to TAR because it achieves the efficiency and reliability of advanced analytics without the constraints of what Tredennick calls, "TAR 1.0.”
Comments
You can follow this conversation by subscribing to the comment feed for this post.