Encoding the expert’s mind
-
Sample.
Zoom selects samples of documents from the collection.
-
Train.
An attorney, expert in the case, tags the sample docs as relevant or not.
-
Iterative.
In a cyclical, self-correcting process, the expert “trains” Zoom to evaluate a document’s relevance.
-
Monitor.
Zoom uses a patented statistical model to continually monitor and analyze training progress.
-
Calculate.
When the training process has optimized, Zoom is able to reliably calculate the relevance score (0 to 100) for each document.
Re-discover e-discovery
-
ECA.
Zoom in on the most relevant documents to make early assessments of case winnability.
-
Culling.
Use Zoom’s Relevance app to find more of the good docs, and less of the bad ones.
-
Prioritized review.
Start with the most relevant documents, then work back.
-
Stratified review.
Divide the collection by relevance. Assign highly relevant documents for in-house review. Assign low scoring documents to outsourced review.
-
Single-pass review.
Use Zoom to create a compact review set, eliminating the need for first-pass review.
-
Review QA.
Cross-match Zoom’s relevance designations against the results of human review. Zoom in on the discrepancies to systemize QA.
The Responsible Choice for Predictive Coding
-
Proven.
Proven in thousands of cases, including DoJ engagements. Proven in TREC. Proven with leading law and consulting firms such as Baker & McKenzie, Sidley Austin, Squire Sanders and KPMG
-
Defensible.
Based on a sound, scientifically valid statistical model. Zoom’s Relevance app is designed for defensibility from Z to M. Including defensible sampling strategies, training methodology, statistical quantification and QA techniques.
-
Setting the standard.
Relevance includes a decision-support environment for review set construction, active learning for accelerated training, graduated relevance scores and support for multi-issues, incremental loads and collaborative training
Zoom in. Find out.
-
Less risk.
-
In ECA, focus on the key documents to allow informed estimates of case risk and winnability
-
Discharge discovery obligations by finding more of the relevant data
-
Prioritize review efforts by focusing on the key data
-
Enable systematic QA of review by cross-matching Relevance versus human review
-
Less cost.
-
Filter non-relevant documents to reduce the review burden
-
Stratify review by assigning high-potential documents for high-grade review, and low-potential documents for low-cost review
-
Address proportionality considerations by using the decision-support environment to determine the size of the review set
-
Eliminate the need for first-pass review
|
|
