Computer-Implemented System And Method For Visually Suggesting Classification For Inclusion-Based Cluster Spines

• US 20140236947A1
• Filed: 04/28/2014
• Published: 08/21/2014
• Est. Priority Date: 07/28/2009
• Status: Active Grant

• Current Assignee: Nuix North America Inc. (Nuix Ltd.)
• Original Assignee: FTI Consulting Incorporated
• Inventors: Knight, William C., Nussbaum, Nicholas I.
• Granted Patent: US 9,542,483 B2
• Time in Patent Office: Days
• US Class Current: 707/737
• CPC Class Codes: G06F 16/287 Visualization; Browsing
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/047 Pattern matching networks; Rete networks
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A computer-implemented system and method for visually suggesting classification for inclusion-based document cluster spines are provided. A set of reference documents each associated with a classification code is designated. A different set of uncoded documents is obtained. One or more of the coded reference documents are combined with a plurality of uncoded documents into a combined document set. The documents in the combined document set are grouped into clusters. The clusters are organized along one or more spines, each spine including a vector. A visual suggestion for assigning one of the classification codes to one of the spines is provided, including visually representing each of the reference concepts in the clusters along that spine.

Claims

1. A computer-implemented system for visually suggesting classification for inclusion-based document cluster spines, comprising the steps of:

a computer processor configured to execute code, comprising;

a designation module configured to designate a set of reference documents each associated with a classification code;

an obtaining module configured to obtain a different set of uncoded documents;

a combination module configured to combine one or more of the coded reference documents with a plurality of uncoded documents into a combined document set;

a grouping module configured to group the documents in the combined document set into clusters;

an organizing module configured to organize the clusters along one or more spines, each spine comprising a vector; and

a suggestion module configured to provide a visual suggestion for assigning one of the classification codes to one of the spines comprising visually representing each of the reference concepts in the clusters along that spine.

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

a presence module configured to provide at least one of a presence and an absence of the documents with each of the classification codes in the clusters along that spine; and

a number module configured to provide a number of the documents with each of the classification codes in the clusters along that spine,

wherein the suggestion includes the number and the presence and the absence.

3. A system according to claim 2, further comprising:

a cluster classification module configured to provide a visual classification suggestion for at least one of the clusters and one or more uncoded documents in that cluster, the suggestion comprising the presence and the absence and the number for that cluster.

4. A system according to claim 1, further comprising:

a selection module configured to receive a user-selection of parameters for defining one or more of sources, custodians, and the classification codes of the reference documents; and

a receipt module configured to receive a user-selection of parameters for defining one or more of commands relating to the reference documents and the uncoded documents, thresholds for the clustering, and automatically assigning one of the classification codes to one of the uncoded documents.

5. A system according to claim 4, wherein the sources comprise those of the reference documents for which the associated classification codes have been verified, those of the reference documents that have been analyzed, and those of the reference documents associated with one of a plurality of document review projects.

6. A system according to claim 1, further comprising:

a compass module configured to provide a compass within which one or more of the clusters organized along the spines are displayed on a display;

a cluster display module configured to display different one or more of the clusters in the compass upon receiving a user command,

wherein the clusters are emphasized when displayed within the compass and deemphasized when displayed outside of the compass.

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

a label module configured to associate a label with each of the spines, each label associated with one or more concepts from the documents in the clusters along that spine; and

a label display module configured to display the labels circumferentially outside of the compass,

wherein the displayed labels do not overlap.

8. A system according to claim 1, wherein the visual representation of one of the reference documents associated with one of the classification codes comprises at least one of a symbol, shape, and color different from the visual representations of the reference documents with the remaining classification codes.

9. A system according to claim 1, further comprising:

a center module configured to identify one of the documents as a center of one of the clusters;

a vector module configured to generate a score vector for the cluster center;

a comparison module configured to compare the score vector for the cluster center to score vectors associated with one or more of the reference documents;

a neighborhood module configured to identify a neighborhood of similar reference documents for the cluster based on the comparison; and

an assignment module configured to assign one of the classification codes to the cluster based on the neighborhood.

10. A system according to claim 9, further comprising:

a distance module configured to determine a distance between the cluster center and the reference documents in the neighborhood; and

a generation module configured to generate the classification code for assignment to the cluster, comprising at least one of;

a closest distance module configured to identify the reference document with the closest distance to the cluster center and assign the classification code of the reference document with the closest distance as the generated classification code for the cluster;

an average distance module configured to calculate an average of the distances between the cluster center and the reference documents associated with each of the classification codes and assign the classification code with the closest average distance as the generated classification code of the cluster; and

a weighted count module configured to count the reference documents in the neighborhood for each of the classification codes, weigh each count based on the distance between the reference documents with the classification code and the cluster center, and assign the classification code with the highest weighted count as the generated classification code of the cluster.

11. A computer-implemented method for visually suggesting classification for inclusion-based document cluster spines, comprising the steps of:

designating a set of reference documents each associated with a classification code;

obtaining a different set of uncoded documents;

combining one or more of the coded reference documents with a plurality of uncoded documents into a combined document set;

grouping the documents in the combined document set into clusters;

organizing the clusters along one or more spines, each spine comprising a vector; and

providing a visual suggestion for assigning one of the classification codes to one of the spines comprising visually representing each of the reference concepts in the clusters along that spine,

wherein the steps are performed by a suitably programmed computer.

12. A method according to claim 11, further comprising:

providing at least one of a presence and an absence of the documents with each of the classification codes in the clusters along that spine; and

providing a number of the documents with each of the classification codes in the clusters along that spine,

wherein the suggestion includes the number and the presence and the absence.

13. A method according to claim 12, further comprising:

providing a visual classification suggestion for at least one of the clusters and one or more uncoded documents in that cluster, the suggestion comprising the presence and the absence and the number for that cluster.

14. A method according to claim 11, further comprising:

receiving a user-selection of parameters for defining one or more of sources, custodians, and the classification codes of the reference documents; and

receiving a user-selection of parameters for defining one or more of commands relating to the reference documents and the uncoded documents, thresholds for the clustering, and automatically assigning one of the classification codes to one of the uncoded documents.

15. A method according to claim 14, wherein the sources comprise those of the reference documents for which the associated classification codes have been verified, those of the reference documents that have been analyzed, and those of the reference documents associated with one of a plurality of document review projects.

16. A method according to claim 11, further comprising:

providing a compass within which one or more of the clusters organized along the spines are displayed on a display;

displaying different one or more of the clusters in the compass upon receiving a user command,

wherein the clusters are emphasized when displayed within the compass and deemphasized when displayed outside of the compass.

17. A method according to claim 16, further comprising:

associating a label with each of the spines, each label associated with one or more concepts from the documents in the clusters along that spine; and

displaying the labels circumferentially outside of the compass,

wherein the displayed labels do not overlap.

18. A method according to claim 11, wherein the visual representation of one of the reference documents associated with one of the classification codes comprises at least one of a symbol, shape, and color different from the visual representations of the reference documents with the remaining classification codes.

19. A method according to claim 11, further comprising:

identifying one of the documents as a center of one of the clusters;

generating a score vector for the cluster center;

comparing the score vector for the cluster center to score vectors associated with one or more of the reference documents;

identifying a neighborhood of similar reference documents for the cluster based on the comparison; and

assigning one of the classification codes to the cluster based on the neighborhood.

20. A method according to claim 19, further comprising:

determining a distance between the cluster center and the reference documents in the neighborhood; and

generating the classification code for assignment to the cluster, comprising at least one of;

identifying the reference document with the closest distance to the cluster center and assigning the classification code of the reference document with the closest distance as the generated classification code for the cluster;

calculating an average of the distances between the cluster center and the reference documents associated with each of the classification codes and assigning the classification code with the closest average distance as the generated classification code of the cluster; and

counting the reference documents in the neighborhood for each of the classification codes, weighing each count based on the distance between the reference documents with the classification code and the cluster center, and assigning the classification code with the highest weighted count as the generated classification code of the cluster.