Workspace setup

A Sample-Based Learning project is an object in which you define the document set to use as the basis for categorizing a larger set of documents, as well as the responsiveness fields used by reviewers to code the set of documents that you want Relativity to recognize so that it can correctly categorize responsive and non-responsive documents going forward using that set as the basis . Once you create a project, you run rounds on it to further specify how documents should be categorized.

The page contains the following sections:

Required workspace components

A Sample-Based Learning project uses the following components, so you'll need to create them before you can create a project. Even if these items already exist in the workspace, you may want to create a new instance of each specifically for your new project.

Before reviewing for a round, all the document coding fields set below during project creation should be available on a document layout (e.g., Designation, Example Responsiveness). See Layouts for more information .

Note: You cannot use Family propagation in a Sample-Based Learning project. Furthermore, if you use document skip in a Sample-Based Learning project and you intend to code the entire sample set, you must check to make sure that all sample set documents get coded.

The following graphic depicts the components required to save a new project. If any of the items below are missing from your workspace, you won't be able to create and save the project.

(Click to expand)

RAR project required component diagram

Analytics index

You must have an Analytics index that includes the documents to be used in the project.

See Creating an Analytics index for more information.

Note: The Analytics index you use for your project must be active and have queries enabled for your project to function properly.

Fields

Necessary fields for setting up and running a project include:
  • Designation field – this single choice field is limited to two choices, for example, Responsive and Not Responsive. You will get an error if you attempt to create a project that has too many choices on this field, or if no choices have been created.
  • Use as Example field – this Yes/No field is used by reviewers to designate a document as an example to use for categorizing other documents.
  • Text Excerpt field - this long text field holds the text excerpts you apply to coding layouts during manual review (if you are running a project that will use text excerpting). This enables reviewers to select the relevant text during document coding to enhance the training of the system by including the relevant portion of the document as an example.
  • Note: If you excerpt text and save it in the field for a document, that text is always an example regardless of the Use as Example selection. To include the entire document and the text excerpt as two separate examples, select Use as Example. Clear the Use as Example checkbox to only include the text excerpt as an example.

  • Key Issues field (Optional) - Single choice field used for issues (if you are running a project that will categorize documents for key issues). See Modifying issue importance for more information on how to weight issue importance.
  • Key Issues Text Excerpt field (Optional) - Long text field used for text excerpting (if you are running a project that will use text excerpting). This enables reviewers to select the relevant text during document coding to enhance the training of the system by including the relevant portion of the document as an example. See Modifying issue importance for more information on how to weight issue importance.

Saved search

Project saved search - includes documents to be categorized that do not include documents that cannot be categorized (e.g., documents must have text). See Saved search for more information.

Sample-Based Learning project checklist

You may want to use the following checklist to make sure that you have all the items needed to run a Sample-Based Learning project:

Sample-Based Learning componentCompleted?
Install the Assisted Review application 
Complete Analytics index build 

Set up the following required project fields:

  • Designation field - single choice
    • Choices: Responsive, Not Responsive
  • Use as Example field - Yes/No
  • Text Excerpt field - Long text
 

(Optional) Set up the following fields (when categorizing for issues):

  • Key Issues field - Single choice
  • Key Issues Text Excerpt field - Long text
 

Create any necessary reviewer views

 

Make all the document coding fields set during project creation available on a document layout.

  • Designation, Use as example, Text excerpt field, etc.
 

(Optional) Perform the following step especially when categorizing for issues:

  • Cluster all documents in a Sample-Based Learning universe
 

Perform document culling to remove documents that won't work well with Analytics indexes:

  • Eliminate documents that do not contain text
  • Run the extracted text size script to review which documents are extremely low on text
  • If documents are clustered, pull UNCLUSTERED documents from the project.

(Optional) You can also use the following additional operations to cull documents that may not be important to the case:

  • Email threading
  • Keyword culling
  • Search terms reports
 

Create any necessary saved searches:

  • Project saved search - documents to be categorized that do not include documents that cannot be categorized (e.g., documents must have text).