US20180173788A1 - System And Method For Providing Inclusion-Based Electronically Stored Information Item Classification Suggestions With The Aid Of A Digital Computer - Google Patents

Abstract

A system and method for providing inclusion-based electronically stored information item classification suggestions with the aid of a digital computer are provided. A set of reference electronically stored information items is designated. A set of uncoded electronically stored information items is obtained. One or more of the coded reference electronically stored information items are combined with the set. A score vector is generated for each of the combined uncoded and reference electronically stored information items. The combined uncoded electronically stored information items are grouped into a plurality of clusters. The score vectors are used to generate a suggestion for assigning one of the classification codes. The visual representations of each of the classification codes of the reference electronically stored information items are displayed as a further suggestion for assigning one of the classification codes. An assignment of one of the classification codes is received.

Description

CROSS-REFERENCE TO RELATED APPLICATION
• This patent application is a continuation of commonly-assigned U.S. patent application Ser. No. 15/401,450, filed Jan. 9, 2017, pending; which is a continuation of U.S. Pat. No. 9,542,483, issued on Jan. 10, 2017; which is a continuation of U.S. Pat. No. 8,713,018, issued Apr. 29, 2014, which claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application, Ser. No. 61/229,216, filed Jul. 28, 2009, and U.S. Provisional Patent Application, Ser. No. 61/236,490, filed Aug. 24, 2009, the disclosures of which are incorporated by reference.
FIELD
• This application relates in general to using electronically stored information as a reference point and, in particular, to a system and method for providing inclusion-based electronically stored information item classification suggestions with the aid of a digital computer.
BACKGROUND
• Historically, document review during the discovery phase of litigation and for other types of legal matters, such as due diligence and regulatory compliance, have been conducted manually. During document review, individual reviewers, generally licensed attorneys, are assigned sets of documents for coding. A reviewer must carefully study each document and categorize the document by assigning a code or other marker from a set of descriptive classifications, such as “privileged,” “responsive,” and “non-responsive.” The classifications can affect the disposition of each document, including admissibility into evidence.
• During discovery, document review can potentially affect the outcome of the underlying legal matter, so consistent and accurate results are crucial. Manual document review is tedious and time-consuming. Marking documents is solely at the discretion of each reviewer and inconsistent results may occur due to misunderstanding, time pressures, fatigue, or other factors. A large volume of documents reviewed, often with only limited time, can create a loss of mental focus and a loss of purpose for the resultant classification. Each new reviewer also faces a steep learning curve to become familiar with the legal matter, classification categories, and review techniques.
• Currently, with the increasingly widespread movement to electronically stored information (ESI), manual document review is no longer practicable. The often exponential growth of ESI exceeds the bounds reasonable for conventional manual human document review and underscores the need for computer-assisted ESI review tools.
• Conventional ESI review tools have proven inadequate to providing efficient, accurate, and consistent results. For example, DiscoverReady LLC, a Delaware limited liability company, custom programs ESI review tools, which conduct semi-automated document review through multiple passes over a document set in ESI form. During the first pass, documents are grouped by category and basic codes are assigned. Subsequent passes refine and further assign codings. Multiple pass review requires a priori project-specific knowledge engineering, which is only useful for the single project, thereby losing the benefit of any inferred knowledge or know-how for use in other review projects.
• Thus, there remains a need for a system and method for increasing the efficiency of document review that bootstraps knowledge gained from other reviews while ultimately ensuring independent reviewer discretion.
SUMMARY
• Document review efficiency can be increased by identifying relationships between reference ESI and uncoded ESI and providing a suggestion for classification based on the relationships. The reference ESI and uncoded ESI are clustered based on a similarity of the ESI. The clusters and the relationship between the uncoded ESI and reference ESI within the clusters are visually depicted. The visual relationship of the uncoded ESI and reference ESI provide a suggestion regarding classification for the uncoded ESI.
• In one embodiment, a system and method for providing inclusion-based electronically stored information item classification suggestions with the aid of a digital computer are provided. A set of reference electronically stored information items is designated, each associated with one of a plurality of classification codes and a visual representation of that classification code. A set of uncoded electronically stored information items is obtained. One or more of the coded reference electronically stored information items are combined with the set of the uncoded electronically stored information items. A score vector is generated for each of the combined uncoded and reference electronically stored information items. The combined uncoded electronically stored information items are grouped into a plurality of clusters using the score vectors. The score vectors of one or more of the uncoded electronically stored information items in one or more of the clusters and the reference electronically stored information items in those clusters are used to generate and provide to user a suggestion for assigning one of the classification codes to the one or more the uncoded electronically stored information items in the one or more of the clusters. The visual representations of each of the classification codes of the reference electronically stored information items in the one or more clusters are displayed as a further suggestion for assigning one of the classification codes to each of the uncoded information in one or more of the clusters. An assignment of one of the classification codes for one or more of the uncoded electronically stored information items for which the suggestion was provided and the further suggestion was displayed is received.
• Still other embodiments of the present invention will become readily apparent to those skilled in the art from the following detailed description, wherein are described embodiments by way of illustrating the best mode contemplated for carrying out the invention. As will be realized, the invention is capable of other and different embodiments and its several details are capable of modifications in various obvious respects, all without departing from the spirit and the scope of the present invention. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a block diagram showing a system for displaying relationships between electronically stored information to provide classification suggestions via inclusion, in accordance with one embodiment.
• FIG. 2 is a process flow diagram showing a method for displaying relationships between electronically stored information to provide classification suggestions via inclusion, in accordance with one embodiment.
• FIG. 3 is a block diagram showing, by way of example, measures for selecting reference document subsets for use in the method ofFIG. 2.
• FIG. 4 is a process flow diagram showing, by way of example, a method for forming clusters for use in the method ofFIG. 2.
• FIG. 5 is a screenshot showing, by way of example, a visual display of reference documents in relation to uncoded documents.
• FIG. 6A is a block diagram showing, by way of example, a cluster with “privileged” reference documents and uncoded documents.
• FIG. 6B is a block diagram showing, by way of example, a cluster with “non-responsive” reference documents and uncoded documents.
• FIG. 6C is a block diagram showing, by way of example, a cluster with uncoded documents and a combination of differently classified reference documents.
• FIG. 7 is a process flow diagram showing, by way of example, a method for classifying uncoded documents for use in the method ofFIG. 2.
• FIG. 8 is a screenshot showing, by way of example, a reference options dialogue box for entering user preferences for clustering documents.
DETAILED DESCRIPTION
• The ever-increasing volume of ESI underlies the need for automating document review for improved consistency and throughput. Previously coded ESI, known as reference ESI, offer knowledge gleaned from earlier work in similar legal projects, as well as a reference point for classifying uncoded ESI.
Providing Suggestions Using Reference Documents
• Reference ESI is previously classified by content and can be used to influence classification of uncoded, that is unclassified, ESI. Specifically, relationships between the uncoded ESI and the reference ESI can be visually depicted to provide suggestions, for instance to a human reviewer, for classifying the visually-proximal uncoded ESI.
• Complete ESI review requires a support environment within which classification can be performed.FIG. 1 is a block diagram showing asystem10 for displaying relationships between electronically stored information to provide classification suggestions via inclusion, in accordance with one embodiment. By way of illustration, thesystem10 operates in a distributed computing environment, which includes a plurality of heterogeneous systems and ESI sources. Henceforth, a single item of ESI will be referenced as a “document,” although ESI can include other forms of non-document data, as described infra. Abackend server11 is coupled to astorage device13, which storesdocuments14a, such as uncoded documents, in the form of structured or unstructured data, adatabase30 for maintaining information about the documents, and alookup database38 for storing many-to-many mappings39 between documents and document features, such as concepts. Thestorage device13 also storesreference documents14b, which can provide a training set of trusted and known results for use in guiding ESI classification. The reference documents14bare each associated with an assigned classification code and considered as classified or coded. Hereinafter, the terms “classified” and “coded” are used interchangeably with the same intended meaning, unless otherwise indicated. A set of reference documents can be hand-selected or automatically selected through guided review, which is further discussed below. Additionally, the set of reference documents can be predetermined or can be generated dynamically, as uncoded documents are classified and subsequently added to the set of reference documents.
• Thebackend server11 is coupled to anintranetwork21 and executes aworkbench suite31 for providing a user interface framework for automated document management, processing, analysis, and classification. In a further embodiment, thebackend server11 can be accessed via aninternetwork22. Theworkbench software suite31 includes adocument mapper32 that includes aclustering engine33,similarity searcher34,classifier35, anddisplay generator36. Other workbench suite modules are possible.
• Theclustering engine33 performs efficient document scoring and clustering of uncoded documents and reference documents, such as described in commonly-assigned U.S. Pat. No. 7,610,313, the disclosure of which is incorporated by reference. Clusters ofuncoded documents14aandreference documents14bare formed and organized along vectors, known as spines, based on a similarity of the clusters. The similarity can be expressed in terms of distance. Document clustering is further discussed below with reference toFIG. 4. Theclassifier35 provides a machine-generated suggestion and confidence level for classification of selecteduncoded documents14b, clusters, or spines, as further described below with reference toFIG. 7.
• Thedisplay generator36 arranges the clusters and spines in thematic relationships in a two-dimensional visual display space, as further described below beginning with reference toFIG. 2. Once generated, the visual display space is transmitted to awork client12 by thebackend server11 via thedocument mapper32 for presenting to a reviewer on adisplay37. The reviewer can include an individual person who is assigned to review and classify one or more uncoded documents by designating a code. Hereinafter, the terms “reviewer” and “custodian” are used interchangeably with the same intended meaning, unless otherwise indicated. Other types of reviewers are possible, including machine-implemented reviewers.
• Thedocument mapper32 operates onuncoded documents14a, which can be retrieved from thestorage13, as well as from a plurality of local and remote sources. As well, the local and remote sources can also store the reference documents14b. The local sources includedocuments17 maintained in astorage device16 coupled to alocal server15 anddocuments20 maintained in astorage device19 coupled to alocal client18. Thelocal server15 andlocal client18 are interconnected to thebackend server11 and thework client12 over anintranetwork21. In addition, thedocument mapper32 can identify and retrieve documents from remote sources over aninternetwork22, including the Internet, through agateway23 interfaced to theintranetwork21. The remote sources includedocuments26 maintained in astorage device25 coupled to aremote server24 anddocuments29 maintained in astorage device28 coupled to aremote client27. Other document sources, either local or remote, are possible.
• Theindividual documents14a,14b,17,20,26,29 include all forms and types of structured and unstructured ESI, including electronic message stores, word processing documents, electronic mail (email) folders, Web pages, and graphical or multimedia data. Notwithstanding, the documents could be in the form of structurally organized data, such as stored in a spreadsheet or database.
• In one embodiment, theindividual documents14a,14b,17,20,26,29 include electronic message folders storing email and attachments, such as maintained by the Outlook and Outlook Express products, licensed by Microsoft Corporation, Redmond, Wash. The database can be an SQL-based relational database, such as the Oracle database management system,Release 8, licensed by Oracle Corporation, Redwood Shores, Calif.
• Theindividual documents17,20,26,29 can be designated and stored as uncoded documents or reference documents. One or more of the uncoded documents can be selected for a document review project and stored as a document corpus, as described infra. The reference documents are initially uncoded documents that can be selected from the corpus or other source of uncoded documents, and subsequently classified. The reference documents can assist in providing suggestions for classification of the remaining uncoded documents in the corpus based on visual relationships between the uncoded documents and reference documents. In a further embodiment, the reference documents can provide suggestions for classifying uncoded documents in a different corpus. In yet a further embodiment, the reference documents can be used as a training set to form machine-generated suggestions for classifying uncoded documents, as further described below with reference toFIG. 8.
• The document corpus for a document review project can be divided into subsets of uncoded documents, which are each provided to a particular reviewer as an assignment. To maintain consistency, the same classification codes can be used across all assignments in the document review project. Alternatively, the classification codes can be different for each assignment. The classification codes can be determined using taxonomy generation, during which a list of classification codes can be provided by a reviewer or determined automatically. For purposes of legal discovery, the list of classification codes can include “privileged,” “responsive,” or “non-responsive;” however, other classification codes are possible. A “privileged” document contains information that is protected by a privilege, meaning that the document should not be disclosed or “produced” to an opposing party. Disclosing a “privileged” document can result in unintentional waivers of the subject matter disclosed. A “responsive” document contains information that is related to a legal matter on which the document review project is based and a “non-responsive” document includes information that is not related to the legal matter.
• Thesystem10 includes individual computer systems, such as thebackend server11,work server12,server15,client18,remote server24 andremote client27. The individual computer systems are general purpose, programmed digital computing devices consisting of a central processing unit (CPU), random access memory (RAM), non-volatile secondary storage, such as a hard drive or CD ROM drive, network interfaces, and peripheral devices, including user interfacing means, such as a keyboard and display. The various implementations of the source code and object and byte codes can be held on a computer-readable storage medium, such as a floppy disk, hard drive, digital video disk (DVD), random access memory (RAM), read-only memory (ROM) and similar storage mediums. For example, program code, including software programs, and data are loaded into the RAM for execution and processing by the CPU and results are generated for display, output, transmittal, or storage.
• Identifying relationships between the reference documents and uncoded documents includes clustering.FIG. 2 is a process flow diagram showing amethod40 for displaying relationships between electronically stored information to provide classification suggestions via inclusion, in accordance with one embodiment. A subset of reference documents is identified and selected (block41) from a representative set of reference documents. The subset of reference documents can be predefined, arbitrary, or specifically selected, as discussed further below with reference toFIG. 3. Upon identification, the reference document subset is grouped with uncoded documents (block42). The uncoded documents can include all uncoded documents in an assignment or in a corpus. The grouped documents, including uncoded and reference documents are organized into clusters (block43). Clustering of the documents is discussed further below with reference toFIG. 4.
• Once formed, the clusters can be displayed to visually depict relationships (block44) between the uncoded documents and the reference documents. The relationships can provide a suggestion, which can be used by an individual reviewer for classifying one or more of the uncoded documents, clusters, or spines. Based on the relationships, the reviewer can classify the uncoded documents, clusters, or spines by assigning a classification code, which can represent a relevancy of the uncoded document to the document review project. Further, machine classification can provide a suggestion for classification, including a classification code, based on a calculated confidence level (block45). Classifying uncoded documents is further discussed below with reference toFIG. 7.
• Identifying a Set and Subset of Reference Documents
• Prior to clustering, the uncoded documents and reference documents are obtained. The reference documents used for clustering can include a particular subset of reference documents, which are selected from a general set of reference documents. Alternatively, the entire set of reference documents can be clustered with the uncoded documents. The set of reference documents is representative of the document corpus for a document review project in which data organization or classification is desired. The reference document set can be previously defined and maintained for related document review projects or can be specifically generated for each review project. A predefined reference set provides knowledge previously obtained during the related document review project to increase efficiency, accuracy, and consistency. Reference sets newly generated for each review project can include arbitrary or customized reference sets that are determined by a reviewer or a machine.
• The set of reference documents can be generated during guided review, which assists a reviewer in building a reference document set. During guided review, the uncoded documents that are dissimilar to the other uncoded documents are identified based on a similarity threshold. Other methods for determining dissimilarity are possible. Identifying a set of dissimilar documents provides a group of uncoded documents that is representative of the corpus for the document review project. Each identified dissimilar document is then classified by assigning a particular classification code based on the content of the document to collectively generate a set of reference documents. Guided review can be performed by a reviewer, a machine, or a combination of the reviewer and machine.
• Other methods for generating a reference document set for a document review project using guided review are possible, including clustering. For example, a set of uncoded documents to be classified is clustered, as described in commonly-assigned U.S. Pat. No. 7,610,313, the disclosure of which is incorporated by reference. A plurality of the clustered uncoded documents are selected based on selection criteria, such as cluster centers or sample clusters. The cluster centers can be used to identify uncoded documents in a cluster that are most similar or dissimilar to the cluster center. The identified uncoded documents are then selected for classification by assigning classification codes. After classification, the documents represent a reference set. In a further embodiment, sample clusters can be used to generate a reference document set by selecting one or more sample clusters based on cluster relation criteria, such as size, content, similarity, or dissimilarity. The uncoded documents in the selected sample clusters are then assigned classification codes. The classified documents represent a document reference set for the document review project. Other methods for selecting documents for use as a reference set are possible.
• Once generated, a subset of reference documents is selected from the reference document set for clustering with uncoded documents.FIG. 3 is a block diagram showing, by way of example, measures50 for selecting reference document subsets51 for use in the method ofFIG. 2. Areference document subset51 includes one or more reference documents selected from a set of reference documents associated with a document review project for use in clustering with uncoded documents. The reference document subset can be predefined52, customized54, selected arbitrarily53, or based on similarity55.
• A subset ofpredefined reference documents52 can be selected from a reference set, which is associated with another document review project that is related to the current document review project. Anarbitrary reference subset53 includes reference documents randomly selected from a reference set, which can be predefined or newly generated for the current document review project or a related document review project. A customizedreference subset54 includes reference documents specifically selected from a current or related reference set based on criteria, such as reviewer preference, classification category, document source, content, and review project. Other criteria are possible. The number of reference documents in a subset can be determined automatically or by a reviewer based on reference factors, such as a size of the document review project, an average size of the assignments, types of classification codes, and a number of reference documents associated with each classification code. Other reference factors are possible. In a further embodiment, the reference document subset can include more than one occurrence of a reference document. Other types of reference document subsets and methods for selecting the reference document subsets are possible.
• Forming Clusters
• Once identified, the reference document subset can be used for clustering with uncoded documents from a corpus associated with a particular document review project. The corpus of uncoded documents for a review project can be divided into assignments using assignment criteria, such as custodian or source of the uncoded document, content, document type, and date. Other criteria are possible. In one embodiment, each assignment is assigned to an individual reviewer for analysis. The assignments can be separately clustered with the reference document subset or alternatively, all of the uncoded documents in the corpus can be clustered with the reference document subset. The content of each uncoded document within the corpus can be converted into a set of tokens, which are word-level or character-level n-grams, raw terms, concepts, or entities. Other tokens are possible.
• An n-gram is a predetermined number of items selected from a source. The items can include syllables, letters, or words, as well as other items. A raw term is a term that has not been processed or manipulated. Concepts typically include nouns and noun phrases obtained through part-of-speech tagging that have a common semantic meaning. Entities further refine nouns and noun phrases into people, places, and things, such as meetings, animals, relationships, and various other objects. Entities can be extracted using entity extraction techniques known in the field. Clustering of the uncoded documents can be based on cluster criteria, such as the similarity of tokens, including n-grams, raw terms, concepts, entities, email addresses, or other metadata.
• Clustering provides groupings of related uncoded documents and reference documents.FIG. 4 is a flow diagram showing a routine60 for forming clusters for use in themethod40 ofFIG. 2. The purpose of this routine is to use score vectors associated with the documents, including uncoded and reference documents, to form clusters based on relative similarity. Hereinafter, the term “document” is intended to include uncoded documents and reference documents selected for clustering, unless otherwise indicated. The score vector associated with each document includes a set of paired values for tokens identified in that document and weights, which are based on scores. The score vector is generated by scoring the tokens extracted from each uncoded document and reference document, as described in commonly-assigned U.S. Pat. No. 7,610,313, the disclosure of which is incorporated by reference.
• As an initial step for generating score vectors, each token within a document is individually scored. Next, a normalized score vector is created for the document by identifying paired values, consisting of a token occurring in that document and the scores for that token. The paired values are ordered along a vector to generate the score vector. The paired values can be ordered based on the tokens, including concept or frequency, as well as other factors. For example, assume a normalized score vector for a first document A is {right arrow over (S)}_A={(5, 0.5), (120, 0.75)} and a normalized score vector for another document B is {right arrow over (S)}_B={(3, 0.4), (5, 0.75), (47, 0.15)}. Document A has scores corresponding to tokens ‘5’ and ‘120’ and Document B has scores corresponding to tokens ‘3,’ ‘5’ and ‘47.’ Thus, these documents only have token ‘5’ in common. Once generated, the score vectors can be compared to determine similarity or dissimilarity between the corresponding documents during clustering.
• The routine for forming clusters of documents, including uncoded documents and reference documents, proceeds in two phases. During the first phase (blocks63-68), the documents are evaluated to identify a set of seed documents, which can be used to form new clusters. During the second phase (blocks70-76), any documents not previously placed are evaluated and grouped into the existing clusters based on a best-fit criterion.
• Initially, a single cluster is generated with one or more documents as seed documents and additional clusters of documents are added, if necessary. Each cluster is represented by a cluster center that is associated with a score vector, which is representative of the tokens in all the documents for that cluster. In the following discussion relating toFIG. 4, the tokens include concepts. However, other tokens are possible, as described supra. The cluster center score vector can be generated by comparing the score vectors for the individual documents in the cluster and identifying the most common concepts shared by the documents. The most common concepts and associated weights are ordered along the cluster center score vector. Cluster centers and thus, cluster center score vectors may continually change due to the addition and removal of documents during clustering.
• During clustering, the documents are identified (block61) and ordered by length (block62). The documents can include all reference documents in a subset and one or more assignments of uncoded documents. Each document is then processed in an iterative processing loop (blocks63-68) as follows. The similarity between each document and a center of each cluster is determined (block64) as the cosine (cos) σ of the score vectors for the document and cluster being compared. The cos σ provides a measure of relative similarity or dissimilarity between tokens, including the concepts, in the documents and is equivalent to the inner products between the score vectors for the document and cluster center.
• In the described embodiment, the cos σ is calculated in accordance with the equation:
• $\cos {\sigma }_{\mathrm{AB}}=\frac{〈{\overrightarrow{S}}_A·{\overrightarrow{S}}_B〉}{\left|{\overrightarrow{S}}_A||{\overrightarrow{S}}_B\right|}$
• where_{cos σAB}comprises the similarity metric between document A and cluster center B, {right arrow over (S)}_Acomprises a score vector for the document A, and {right arrow over (S)}_Bcomprises a score vector for the cluster center B. Other forms of determining similarity using a distance metric are feasible, as would be recognized by one skilled in the art. An example includes using Euclidean distance.
• Only those documents that are sufficiently distinct from all cluster centers (block65) are selected as seed documents for forming new clusters (block66). If the document being compared is not sufficiently distinct (block65), the document is then grouped into a cluster with the most similar cluster center (block67). Processing continues with the next document (block68).
• In the second phase, each document not previously placed is iteratively processed in an iterative processing loop (blocks70-76) as follows. Again, the similarity between each remaining document and each of the cluster centers is determined based on a distance (block71), such as the cos σ of the normalized score vectors for each of the remaining documents and the cluster centers. A best fit between a remaining document and a cluster center can be found subject to a minimum fit criterion (block72). In the described embodiment, a minimum fit criterion of 0.25 is used, although other minimum fit criteria could be used. If a best fit is found (block73), the remaining document is grouped into the cluster having the best fit (block75). Otherwise, the remaining document is grouped into a miscellaneous cluster (block74). Processing continues with the next remaining document (block76). Finally, a dynamic threshold can be applied to each cluster (block77) to evaluate and strengthen document membership in a particular cluster. The dynamic threshold is applied based on a cluster-by-cluster basis, as described in commonly-assigned U.S. Pat. No. 7,610,313, the disclosure of which is incorporated by reference. The routine then returns. Other methods and processes for forming clusters are possible.
• Displaying the Reference Documents
• Once formed, the clusters of documents can be can be organized to generate spines of thematically related clusters, as described in commonly-assigned U.S. Pat. No. 7,271,804, the disclosure of which is incorporated by reference. Each spine includes those clusters that share one or more tokens, such as concepts, which are placed along a vector. Also, the cluster spines can be positioned in relation to other cluster spines based on a theme shared by those cluster spines, as described in commonly-assigned U.S. Pat. No. 7,610,313, the disclosure of which is incorporated by reference. Each theme can include one or more concepts defining a semantic meaning. Organizing the clusters into spines and groups of cluster spines provides an individual reviewer with a display that presents the documents according to a theme while maximizing the number of relationships depicted between the documents.
• FIG. 5 is ascreenshot80 showing, by way of example, avisual display81 ofreference documents85 in relation touncoded documents84.Clusters83 can be located along a spine, which is a straight vector, based on a similarity of thedocuments84,85 in theclusters83. Eachcluster83 is represented by a circle; however, other shapes, such as squares, rectangles, and triangles are possible, as described in U.S. Pat. No. 6,888,548, the disclosure of which is incorporated by reference. Theuncoded documents84 are each represented by a smaller circle within theclusters83, while the reference documents85 are each represented by a circle having a diamond shape within the boundaries of the circle. The reference documents85 can be further represented by their assigned classification code. The classification codes can include “privileged,” “responsive,” and “non-responsive” codes, as well as other codes. Each group of reference documents associated with a particular classification code can be identified by a different color. For instance, “privileged” reference documents can be colored blue, while “non-responsive” reference documents are red and “responsive” reference documents are green. In a further embodiment, the reference documents for different classification codes can include different symbols. For example, “privileged” reference documents can be represented by a circle with an “X” in the center, while “non-responsive” reference documents can include a circle with striped lines and “responsive” reference documents can include a circle with dashed lines. Other classification representations for the reference documents are possible. Eachcluster spine86 is represented as a straight vector along which the clusters are placed.
• Thedisplay81 can be manipulated by an individual reviewer via acompass82, which enables the reviewer to navigate, explore, and search theclusters83 andspines86 appearing within thecompass82, as further described in commonly-assigned U.S. Pat. No. 7,356,777, the disclosure of which is incorporated by reference. Visually, thecompass82 emphasizesclusters83 located within thecompass82, while deemphasizingclusters83 appearing outside of thecompass82.
• Spine labels89 appear outside of thecompass82 at an end of eachcluster spine86 to connect the outermost cluster of acluster spine86 to the closest point along the periphery of thecompass82. In one embodiment, the spine labels89 are placed without overlap and circumferentially around thecompass82. Eachspine label89 corresponds to one or more concepts that most closely describe thecluster spines86 appearing within thecompass82. Additionally, the cluster concepts for each of the spine labels89 can appear in a concepts list (not shown) also provided in the display.Toolbar buttons87 located at the top of thedisplay81 enable a user to execute specific commands for the composition of the spine groups displayed. A set of pull downmenus88 provide further control over the placement and manipulation ofclusters83 andcluster spines86 within thedisplay81. Other types of controls and functions are possible.
• A document guide90 can be placed within thedisplay81. The document guide90 can include a “Selected” field, a “Search Results” field, and details regarding the numbers of uncoded documents and reference documents provided in the display. The number of uncoded documents includes all uncoded documents selected for clustering, such as within a corpus of uncoded documents for a review project or within an assignment. The number of reference documents includes the reference document subset selected for clustering. The “Selected” field in the document guide90 provides a number of documents within one or more clusters selected by the reviewer. The reviewer can select a cluster by “double clicking” the visual representation of that cluster using a mouse. The “Search Results” field provides a number of uncoded documents and reference documents that include a particular search term identified by the reviewer in a search query box92.
• In one embodiment, agarbage can91 is provided to remove tokens, such as cluster concepts, from consideration in the current set ofclusters83. Removed cluster concepts prevent those concepts from affecting future clustering, as may occur when a reviewer considers a concept irrelevant to theclusters83.
• Thedisplay81 provides a visual representation of the relationships between thematically-related documents, including the uncoded documents and reference documents. The uncoded documents and reference documents located within a cluster or spine can be compared based on characteristics, such as the assigned classification codes of the reference documents, a number of reference documents associated with each classification code, and a number of different classification codes to identify relationships between the uncoded documents and reference documents. The reviewer can use the displayed relationships as suggestions for classifying the uncoded documents. For example,FIG. 6A is a block diagram showing, by way of example, acluster93 with “privileged” reference documents95 and uncoded documents94. Thecluster93 includes nine uncoded documents94 and threereference documents95. Eachreference document95 is classified as “privileged.” Accordingly, based on the number of “privileged” reference documents95 present in thecluster93, the absence of other classifications of reference documents, and the thematic relationship between the uncoded documents94 and the “privileged” reference documents95, the reviewer may be more inclined to review the uncoded documents94 in thatcluster93 or to classify one or more of the uncoded documents94 as “privileged” without review.
• Alternatively, the three reference documents can be classified as “non-responsive,” instead of “privileged” as in the previous example.FIG. 6B is a block diagram showing, by way of example, acluster96 with “non-responsive” reference documents97 and uncoded documents94. Thecluster96 includes nine uncoded documents94 and three “non-responsive” documents97. Since the uncoded documents94 in the cluster are thematically related to the “non-responsive” reference documents97, the reviewer may wish to assign a “non-responsive” code to one or more of the uncoded documents94 without review, as they are most likely not relevant to the legal matter associated with the document review project. In making a decision to assign a code, such as “non-responsive,” the reviewer can consider the number of “non-responsive” reference documents in the cluster, the presence or absence of other reference document classification codes, and the thematic relationship between the “non-responsive” reference documents and the uncoded documents. Thus, the presence of the three “non-responsive” reference documents97 in the cluster provides a suggestion that the uncoded documents94 may also be “non-responsive.” Further, thelabel89 associated with thespine86 upon which the cluster is located can also be used to influence a suggestion.
• A further example can include a cluster with combination of “privileged” and “non-responsive” reference documents. For example,FIG. 6C is a block diagram showing, by way of example, acluster98 with uncoded documents94 and a combination of differently classifiedreference documents95,97. Thecluster98 can include one “privileged”reference document95, two “non-responsive” reference documents97, and nine uncoded documents94. The “privileged”95 and “non-responsive”97 reference documents can be distinguished by different colors or shape, as well as other identifiers. The combination of “privileged”95 and “non-responsive”97 reference documents within thecluster98 can suggest to a reviewer that the uncoded reference documents94 should be reviewed before classification or that one or more uncoded reference documents94 should be classified as “non-responsive” based on the higher number of “non-responsive” reference documents97 in thecluster98. In making a classification decision, the reviewer may consider the number of “privileged” reference documents95 versus the number of “non-responsive” reference documents97, as well as the thematic relationships between the uncoded documents94 and the “privileged”95 and “non-responsive”97 reference documents. Additionally, the reviewer can identify the closest reference document to an uncoded document and assign the classification code of the closest reference document to the uncoded document. Other examples, classification codes, and combinations of classification codes are possible.
• Additionally, the reference documents can also provide suggestions for classifying clusters and spines. The suggestions provided for classifying a cluster can include factors, such as a presence or absence of classified documents with different classification codes within the cluster and a quantity of the classified documents associated with each classification code in the cluster. The classification code assigned to the cluster is representative of the documents in that cluster and can be the same as or different from one or more classified documents within the cluster. Further, the suggestions provided for classifying a spine include factors, such as a presence or absence of classified documents with different classification codes within the clusters located along the spine and a quantity of the classified documents for each classification code. Other suggestions for classifying documents, clusters, and spines are possible.
Classifying Uncoded Documents
• The display of relationships between the uncoded documents and reference documents provides suggestion to an individual reviewer. The suggestions can indicate a need for manual review of the uncoded documents, when review may be unnecessary, and hints for classifying the uncoded documents. Additional information can be generated to assist the reviewer in making classification decisions for the uncoded documents, such as a machine-generated confidence level associated with a suggested classification code, as described in common-assigned U.S. Pat. No. 8,635,225, issued on Jan. 21, 2014, pending, the disclosure of which is incorporated by reference.
• The machine-generated suggestion for classification and associated confidence level can be determined by a classifier.FIG. 7 is a process flow diagram100 showing, by way of example, a method for classifying uncoded documents by a classifier for use in the method ofFIG. 2. An uncoded document is selected from a cluster within a cluster set (block101) and compared to a neighborhood of x-reference documents (block102), also located within the cluster, to identify those reference documents that are most relevant to the selected uncoded document. In a further embodiment, a machine-generated suggestion for classification and an associated confidence level can be provided for a cluster or spine by selecting and comparing the cluster or spine to a neighborhood of x-reference documents determined for the selected cluster or spine.
• The neighborhood of x-reference documents is determined separately for each selected uncoded document and can include one or more reference documents within that cluster. During neighborhood generation, an x number of reference documents is first determined automatically or by an individual reviewer. Next, the x-number of reference documents nearest in distance to the selected uncoded document are identified. Finally, the identified x-number of reference documents are provided as the neighborhood for the selected uncoded document. In a further embodiment, the x-number of reference documents are defined for each classification code, rather than across all classification codes. Once generated, the x-number of reference documents in the neighborhood and the selected uncoded document are analyzed by the classifier to provide a machine-generated classification suggestion (block103). A confidence level for the suggested classification is also provided (block104).
• The analysis of the selected uncoded document and x-number of reference documents can be based on one or more routines performed by the classifier, such as a nearest neighbor (NN) classifier. The routines for determining a suggested classification code include a minimum distance classification measure, also known as closest neighbor, minimum average distance classification measure, maximum count classification measure, and distance weighted maximum count classification measure. The minimum distance classification measure includes identifying a neighbor that is the closest distance to the selected uncoded document and assigning the classification code of the closest neighbor as the suggested classification code for the selected uncoded document. The closest neighbor is determined by comparing the score vectors for the selected uncoded document with each of the x-number of reference documents in the neighborhood as the cos σ to determine a distance metric. The distance metrics for the x-number of reference documents are compared to identify the reference document closest to the selected uncoded document as the closest neighbor.
• The minimum average distance classification measure includes calculating an average distance of the reference documents in a cluster for each classification code. The classification code with the reference documents having the closest average distance to the selected uncoded document is assigned as the suggested classification code. The maximum count classification measure, also known as the voting classification measure, includes counting a number of reference documents within the cluster for each classification code and assigning a count or “vote” to the reference documents based on the assigned classification code. The classification code with the highest number of reference documents or “votes” is assigned to the selected uncoded document as the suggested classification. The distance weighted maximum count classification measure includes identifying a count of all reference documents within the cluster for each classification code and determining a distance between the selected uncoded document and each of the reference documents. Each count assigned to the reference documents is weighted based on the distance of the reference document from the selected uncoded document. The classification code with the highest count, after consideration of the weight, is assigned to the selected uncoded document as the suggested classification.
• The machine-generated classification code is provided for the selected uncoded document with a confidence level, which can be presented as an absolute value or a percentage. Other confidence level measures are possible. The reviewer can use the suggested classification code and confidence level to assign a classification to the selected uncoded document. Alternatively, the x-NN classifier can automatically assign the suggested classification. In one embodiment, the x-NN classifier only assigns an uncoded document with the suggested classification code if the confidence level is above a threshold value, which can be set by the reviewer or the x-NN classifier.
• Classification can also occur on a cluster or spine level. For instance, for cluster classification, a cluster is selected and a score vector for the center of the cluster is determined as described above with reference toFIG. 4. A neighborhood for the selected cluster is determined based on a distance metric. The x-number of reference documents that are closest to the cluster center can be selected for inclusion in the neighborhood, as described above. Each reference document in the selected cluster is associated with a score vector and the distance is determined by comparing the score vector of the cluster center with the score vector of each reference document to determine an x-number of reference documents that are closest to the cluster center. However, other methods for generating a neighborhood are possible. Once determined, one of the classification measures is applied to the neighborhood to determine a suggested classification code and confidence level for the selected cluster.
• During classification, either by an individual reviewer or a machine, the reviewer can retain control over many aspects, such as a source of the reference documents and a number of reference documents to be selected.FIG. 8 is ascreenshot110 showing, by way of example, anoptions dialogue box111 for entering user preferences for clustering and display of the uncoded documents and reference documents. Thedialogue box111 can be accessed via a pull-down menu as described above with respect toFIG. 5. Within thedialogue box111, the reviewer can utilize user-selectable parameters to define areference source112,category filter113, command details114,advanced options115,classifier parameters116, and commands117. Each user-selectable option can include a text box for entry of a user preference or a drop-down menu with predetermined options for selection by the reviewer. Other user-selectable options and displays are possible.
• Thereference source parameter112 allows the reviewer to identify one or more sources of the reference documents. The sources can include all reference documents for which the associated classification has been verified, all reference documents that have been analyzed, and all reference documents in a particular binder. The binder can include reference documents particular to a current document review project or that are related to a prior document review project. Thecategory filter parameter113 allows the reviewer to generate and display the subset of reference documents using only those reference documents associated with a particular classification code. Other options for generating the reference set are possible, including custodian, source, and content. Thecommand parameters114 allow the reviewer to enter instructions regarding actions for the uncoded and reference documents, such as indicating counts of the documents, and display of the documents. Theadvanced option parameters115 allow the reviewer to specify clustering thresholds and classifier parameters. The parameters entered by the user can be compiled ascommand parameters116 and provided in a drop-down menu on a display of the clusters. Other user selectable parameters, options, and actions are possible.
• Providing suggestions for classification has been described in relation to uncoded documents and reference documents; however, in a further embodiment, suggestions can be provided for tokens extracted from the uncoded documents using reference tokens. For example, the uncoded tokens and reference tokens are clustered and displayed to provide classification suggestions based on relationships between the uncoded tokens and similar reference tokens. The uncoded documents can then be classified based on the classified tokens. In one embodiment, the tokens include concepts, n-grams, raw terms, and entities.
• While the invention has been particularly shown and described as referenced to the embodiments thereof, those skilled in the art will understand that the foregoing and other changes in form and detail may be made therein without departing from the spirit and scope.

Claims (20)

What is claimed is:

1. A system for providing inclusion-based electronically stored information item classification suggestions with the aid of a digital computer, comprising:

a non-transitory computer readable storage medium comprising program code; and

a computer processor configured coupled to the storage medium, wherein the processor is configured to execute the program code to perform steps to:

designate a set of reference electronically stored information items each associated with one of a plurality of classification codes and a visual representation of that classification code;

obtain a set of uncoded electronically stored information items;

combine one or more of the reference electronically stored information items with the set of the uncoded electronically stored information items;

generate a score vector for each of the combined uncoded and reference electronically stored information items;

group the combined uncoded electronically stored information items into a plurality of clusters using the score vectors;

use the score vectors of one or more of the uncoded electronically stored information items in one or more of the clusters and the reference electronically stored information items in those clusters to generate and provide to user a suggestion for assigning one of the classification codes to each of the uncoded electronically stored information items in the one or more of the clusters;

display the visual representations of each of the classification codes of the reference electronically stored information items in the one or more clusters as a further suggestion for assigning one of the classification codes to each of the uncoded information in one or more of the clusters; and

receive an assignment of one of the classification codes for one or more of the uncoded electronically stored information items for which the suggestion was provided and the further suggestion was displayed.

2. The system according toclaim 1, the steps further comprising:

generate a neighborhood of reference electronically stored information items for each of the uncoded electronically stored information items in the one or more clusters, wherein the neighborhood is used for generating the suggestion for that uncoded electronically stored information item.

3. The system according toclaim 2, wherein the neighborhood comprises an equal number of the reference electronically stored information items associated with each of the classification codes.

4. The system according toclaim 2, the steps further comprising:

using a distance metric to determine a distance between each of the uncoded electronically stored information items in one of the one or more clusters and each of the reference electronically stored information items in that cluster, wherein the distance is used for the generation of the neighborhood.

5. The system according toclaim 2, the steps further comprising at least one of:

identify the reference electronically stored information item in the neighborhood of one of the electronically stored information items with the closest distance to that uncoded electronically stored information item and assign the classification code of the reference electronically stored information item with the closest distance as the suggested classification code for that uncoded electronically stored information item;

calculate an average of the distances between one of the uncoded electronically stored information items and the reference electronically stored information items in the neighborhood of that electronically stored information item that are associated with each of the classification codes and assign the classification code with the closest average distance as the suggested classification code for that uncoded electronically stored information item; and

count in the neighborhood for one of the uncoded electronically stored information items the reference electronically stored information items for each of the classification codes, weighing each count based on the distance between the reference electronically stored items with the classification code and that uncoded electronically stored information item, and assign the classification code with the highest weighted count as the suggested classification code for that uncoded electronically stored information item.

6. The system according toclaim 1, the steps further comprising:

organize the clusters along a plurality of spines and displaying the spines on a display, wherein the spines are positioned in relation to each other based on a theme shared by one or more of the spines.

7. The system according toclaim 6, the steps further comprising:

receive a user selection of one of the clusters in one of the displayed spines; and

display a number of the reference electronically stored information items and the uncoded electronically stored information items within the selected cluster.

8. The system according toclaim 6, the steps further comprising:

receive a user query comprising a search term; and

display a number of the reference electronically stored information items and the uncoded electronically stored information items that comprise the search results.

9. The system according toclaim 6, the steps further comprising:

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

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.

10. The system according toclaim 9, the steps further comprising:

provide a confidence level for each of the suggestions.

11. A method for providing inclusion-based electronically stored information item classification suggestions with the aid of a digital computer, comprising the steps of:

designating a set of reference electronically stored information items each associated with one of a plurality of classification codes and a visual representation of that classification code;

obtaining a set of uncoded electronically stored information items;

combining one or more of the reference electronically stored information items with the set of the uncoded electronically stored information items;

generating a score vector for each of the combined uncoded and reference electronically stored information items;

grouping the combined uncoded electronically stored information items into a plurality of clusters using the score vectors;

using the score vectors of one or more of the uncoded electronically stored information items in one or more of the clusters and the reference electronically stored information items in those clusters to generate and provide to user a suggestion for assigning one of the classification codes to the one or more of the uncoded electronically stored information items in the one or more of the clusters;

displaying the visual representations of each of the classification codes of the reference electronically stored information items in the one or more clusters as a further suggestion for assigning one of the classification codes to each of the uncoded electronically stored information items in the one or more of the clusters; and

receiving an assignment of one of the classification codes for one or more of the uncoded electronically stored information items for which the suggestion was provided and the further suggestions was displayed,

wherein the steps are performed on a suitably-programmed computer.

12. The method according toclaim 11, further comprising:

generating a neighborhood of reference electronically stored information items for each of the uncoded electronically stored information items in the one or more clusters, wherein the neighborhood is used for generating the suggestion for that uncoded electronically stored information item.

13. The method according toclaim 12, wherein the neighborhood comprises an equal number of the reference electronically stored information items associated with each of the classification codes.

14. The method according toclaim 12, further comprising:

using a distance metric to determine a distance between each of the uncoded electronically stored information items in one of the one or more clusters and each of the reference electronically stored information items in that cluster, wherein the distance is used for the generation of the neighborhood.

15. The method according toclaim 12, further comprising at least one of:

identifying the reference electronically stored information item in the neighborhood of one of the electronically stored information items with the closest distance to that uncoded electronically stored information item and assigning the classification code of the reference electronically stored information item with the closest distance as the suggested classification code for that uncoded electronically stored information item;

calculating an average of the distances between one of the uncoded electronically stored information items and the reference electronically stored information items in the neighborhood of that electronically stored information item that are associated with each of the classification codes and assigning the classification code with the closest average distance as the suggested classification code for that uncoded electronically stored information item; and

counting in the neighborhood for one of the uncoded electronically stored information items the reference electronically stored information items for each of the classification codes, weighing each count based on the distance between the reference electronically stored items with the classification code and that uncoded electronically stored information item, and assigning the classification code with the highest weighted count as the suggested classification code for that uncoded electronically stored information item.

16. The method according toclaim 11, further comprising:

organizing the clusters along a plurality of spines and displaying the spines on a display, wherein the spines are positioned in relation to each other based on a theme shared by one or more of the spines.

17. The method according toclaim 16, further comprising:

receiving a user selection of one of the clusters in one of the displayed spines;

displaying a number of the reference electronically stored information items and the uncoded electronically stored information items within the selected cluster.

18. The method according toclaim 16, further comprising:

receiving a user query comprising a search term; and

displaying a number of the reference electronically stored information items and the uncoded electronically stored information items that comprise the search results.

19. The method according toclaim 16, further comprising:

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

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.

20. The method according toclaim 19, further comprising:

providing a confidence level for each of the suggestions.

Images