US Release 1/26/2018
This release is huge. We are excited and proud to present three major features:
Data Visualizer: All users can explore their documents, filter by various document or review properties, and display results visually.
Predictive coding system update: Your prediction models are more defensible, more visually accessible, and easier to navigate for the results that matter most to you.
Homepage folders and card redesign: Organize your homepage the way you want. Create folders, hide extraneous searches, and share folder contents. Homepage cards have undergone a redesign as well!
In addition to these three major features, we present a host of small improvements as well:
- Searching against a prior search
- Support of native Google Drive file uploads
- Sorting by billable size
- [Organization admin only] Interface changes to dashboard
Read on more for details! If you have questions, comments, or feedback, feel free to contact us at firstname.lastname@example.org.
Data Visualizer provides an overview of the documents in your database by summarizing their characteristics visually. With Data Visualizer, you can explore documents at a glance without the need to review individual documents or create multiple explanatory searches. This is particularly useful in the context of early case assessment.
Every user can access Data Visualizer from the search page or results table. For an overview of Data Visualizer, click on the video below:
Predictive Coding System Update
Long-time users will notice our predictive coding system has a fresh new look.
In addition, the new system makes it easier to create and train new prediction models, provides more guidance on how best to make use of a model's predictions, and makes the evaluation of a model's performance richer and more interactive. We are also releasing a comprehensive Beginner’s Guide to Predictive Coding for a high-level understanding of predictive coding.
Below are more specifics about the exciting upgrades to our predictive coding system:
- Easier model creation: We have changed how models are created. In the past, you would specify the relevant and irrelevant criteria for a model, but thinking through multiple queries and double negatives can be a challenge. Now, you’ll set the criteria for documents considered reviewed—or those within the universe of documents you’d like to be considered for training. You can then specify the subset of those documents that are “relevant.” All documents that are "reviewed” and not “relevant” will be considered “irrelevant” for the model. Learn more about creating a new model.
- Update the model on demand: You can now initiate an immediate update by clicking the “update now” button at the bottom of the predictive coding page.
- Performance metrics at any relevance boundary: With this release, you can select any relevance boundary and see the precision, recall, and F1 values at that boundary. This allows you to easily select a body of documents based on your desired precision and recall levels. Read more information.
- Introduction of holdout sets and historical performance insights: Holdout sets are 5% of the total case documents carved out for the purpose of evaluating the model. Documents in the holdout set are not used to train the model; they are reserved exclusively for evaluating a model’s performance. Because the documents in the holdout set always remain in the holdout set, you can measure historical performance of the model. With this release, you can see how your precision, recall, and F1 scores have changed over time, giving you greater insight into the ROI and progress of training a model. More detailed information.
Visit the predictive coding section of our help center for details on creating, interpreting, and leveraging our predictive coding system.
Homepage Folders and Card Redesign
This release introduces a much-anticipated feature: homepage folders!
You can create folders that can store multiple homepage cards, while viewing and organizing your homepage more efficiently. Share folder contents with your colleagues while managing card permissions. Learn more about homepage folders.
In addition to homepage folders, cards on the homepage have gotten a facelift! Cards will now display not only a colored band but also colored text indicating the card type. This has been introduced to alleviate confusion when multiple card types were located in the same column.
Cards under the Document Sets column on the homepage now show the date of the upload in the lower right-hand corner of the card, including native uploads cards.
Searching Against a Prior Search
We’ve introduced a search term that allows you to search against a prior search. For example, you will now be able to search for documents that were included in one search, but not included in a second search. You can find the Prior Search term in the Review category.
Support of Native Google Drive File Uploads
We now support the ingestion of Google Drive files. You can upload native Google Drive documents such as Docs, Sheets, Slides, or Drawings, through the cloud uploader. The documents are converted to their Microsoft Office equivalents upon upload (for example, a Google Slides document is converted to a Powerpoint, or .pptx, file, and will be searchable under Type: Presentation).
Our full list of acceptable native file types now includes Google Drive files.
Ability to Sort by Billable Size
Just as it sounds, you can now sort document results by billable size! This can help you find the largest documents in a set, guiding culling decisions.
[Organization Admins Only] Interface Changes to Dashboard
The organization admin dashboard has a fresh new look to make the experience better for you and the UI more consistent with the rest of our platform. Functionally, the organization admin page will remain the same.
Organization admin functionality is available to users with annual contracts. Organization admins can create new databases, upload processed data, rename and suspend projects, track project sizes, and much more without needing to contact Everlaw Support.