System and method for generating a reference set for use during document review

• US 8,612,446 B2
• Filed: 08/24/2010
• Issued: 12/17/2013
• Est. Priority Date: 08/24/2009
• Status: Active Grant

• Current Assignee: Nuix North America Inc. (Nuix Ltd.)
• Original Assignee: FTI Consulting Incorporated
• Inventors: Knight, William C., McNee, Sean M., Conwell, John
• Primary Examiner(s): Fleurantin, Jean B
• Application Number:
  US12/862,682
• Publication Number:
  US 20110047156A1
• Time in Patent Office: 1,211 Days
• Field of Search: 707/708, 707/737
• US Class Current: 707/737
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

A system and method for providing generating reference sets for use during document review is provided. A collection of unclassified documents is obtained. Selection criteria are applied to the document collection and those unclassified documents that satisfy the selection criteria are selected as reference set candidates. A classification code is assigned to each reference set candidate. A reference set is formed from the classified reference set candidates. The reference set is quality controlled and shared between one or more users.

Claims

1. A method for generating a reference set for use during document review, using one or more processors, comprising:

obtaining a collection of unclassified documents;

identifying one or more features within each of the unclassified documents;

generating clusters of the features and selecting at least one feature from one or more of the clusters as reference set candidates;

assigning a classification code to each reference set candidate; and

refining the reference set candidates, comprising;

grouping the reference set candidates into further clusters;

selecting at least one of the reference set candidates in one or more of the further clusters as the further reference set candidates;

assigning a classification code to the further reference set candidates; and

forming a reference set from the unclassified documents associated with the further classified reference set candidates; and

wherein all the steps are performed by a suitably programmed computer.

2. A method according to claim 1, further comprising:

refining the reference set by reducing a number of reference set candidates included in the reference set.

3. A method according to claim 1, further comprising:

building a hierarchical tree of the clusters; and

traversing the hierarchical tree to identify the reference set candidates.

4. A system for generating a reference set for use during document review, comprising:

a collection of unclassified documents;

a clustering module to identify one or more features within each of the unclassified documents, to generate clusters of the features, and to select at least one feature from one or more of the clusters as reference set candidates;

a classification module to assign a classification code to each reference set candidate;

an identification module to refine the reference set candidates by grouping the reference set candidates into further clusters, selecting at least one of the reference set candidates in one or more of the further clusters, and assigning a classification code to the further reference set candidates;

a data module to form a reference set from the unclassified documents associated with the further classified reference set candidates; and

a processor to execute the modules.

5. A system according to claim 4, further comprising:

a refinement module to refine the reference set by reducing a number of reference set candidates included in the reference set.

6. A system according to claim 4, further comprising:

building a hierarchical tree of the clusters; and

traversing the hierarchical tree to identify the reference set candidates.