Reveal Help Center

Getting Started: Data Visualizations

Logging into Brainspace

Verify internet browser

  • Brainspace uses advanced web graphics that are ONLY supported in the newer internet browsers. Please ensure you verify that you have either Google Chrome, Firefox, or Microsoft Edge installed on your computer before logging in to Brainspace for the first time.

    Internet Browser

    Browser Version

    Google Chrome

    61 to Current

    Firefox

    57 to Current

    Firefox ESR

    52.4 to Current

    Edge

    41 to Current

    Note

    Please note: Microsoft Internet Explorer and Safari are not supported by Brainspace.

Login to Brainspace

  • Open your internet browser (Microsoft Edge, Google Chrome, Firefox).

  • Type or paste the URL to your Brainspace application into the internet browser address bar.

  • Enter your Brainspace username and password at the Brainspace login prompt.

    image1.png
  • The first time you log in to Brainspace, you will be prompted to change your password by typing the current password and then a new password (twice for confirmation).

  • Once you have changed your password, click on the down arrow in the upper right-hand corner of the screen and then click on Sign Out.

    image2.png
  • Now log back into Brainspace with your username and newly created password

Datasets

A single project or specific set of data to be analyzed. In the legal industry each Dataset represents a specific legal matter where data was collected and ingested into Brainspace to determine what evidence exists to support an accusation of wrongdoing.

Dataset dashboard

A tile view listing of all the Datasets to which you have access based on your user permissions. Each tile represents a single Dataset. Users can filter the Dataset Dashboard by name by entering a partial name of the full name of the Dataset Name into the Filter by dataset name text box at the top of this screen.

image3.png

Dataset Tile

The information on the Dataset Tile includes the Dataset Name, number of documents or records contained within the Dataset, the number of new documents (any documents recently added) and the Dataset Status.

image4.png

Access Dataset

  • To access a Dataset, click on the appropriate Dataset Tile from within the Dataset Dashboard.

  • The Dataset Selection Drop Down menu is located in the upper left-hand corner of the user interface. You can use this drop-down menu to navigate to other Datasets that are in a Running state/status within your Brainspace instance.

    image5.png

Analytics, Notebooks & Supervised Learning Tabs

The Analytics, Notebooks & Classification tabs are located in the upper right-hand corner of the user interface. Please see description below for each tab.

  • Analytics Tab – This menu option provides you access to all of the Brainspace Data Visualizations.

  • Notebooks Tab – This menu option provides access to all of the Notebooks created for the Dataset you are currently working in.

  • Supervised Learning Tab – The Classification tab provides you with access to Brainspace’s Supervised Machine Learning features. Please see section on Classification to learn more about these features.

    image6.png

Data Visualizations

Dashboard

A series of data visualizations that present structured data in several different visual graphs allowing the user to interact with the information and make decisions as to what information is important or apply filters to remove unimportant or non-relevant data from view.

The Dashboard can be accessed by clicking on the Dashboard menu option in the top left-hand corner of the Analytics tab.

image7.png
  • Total Document Count – Total number of documents after any active filters or searches have been applied to the Dataset.

    image8.png
  • Candy Bar – Data visualization that allows you to either include or exclude Originals, Near Duplicates, Exact Duplicates, and/or Not Analyzed documents in your results.

    image9.png
  • Originals – documents in the Dataset that have unique content.

  • Near Duplicates – documents in the Dataset where the text of the document is at least 80% similar to an Original document.

  • Exact Duplicates – documents in the Dataset that are considered exact (100%) text duplicates of an Original document.

  • Not Analyzed – documents that either have no text, very little text, or garbled text which prevented Brainspace from analyzing the document for deduplication or clusterization purposes (see Cluster Wheel section below).

  • Timeline Graph – This data visualization organizes the data by date and time. Any metadata date field can be added to Brainspace in order to view and analyze data across time. By default, the graph charts the data by Year. Users can click on a specific year to then see how the data breaks down within that Year. Users can then click on a specific Month and so on until you drill down to a Minute by Minute view of the data if Time metadata has also been added with Date field during data ingestion into Brainspace. Users can view this graph on any ingested Date field. To view the data using a different Date field, select the appropriate Date from the Date Field Selection drop-down menu (if present).

    image10.png
  • Term Heat Map – This data visualization organizes the data by Date/Time and by Concept or Term. The graph shows you when (Month/Year/Day/Time of Day, etc.) 10 terms heat up or cool off over time. In other words, the graph shows when there is a high concentration of documents that include a specific term for a specific period of time. The lighter shaded cells (in light blue) indicate there are more documents related to a specific term for a specific time period; the lighter the blue shading the more documents exist for that term and time period. The Term Heat Map is tightly integrated with the Timeline Graph so that any selection you make in the Timeline Graph will be immediately reflected in the Term Heat Map.

    image11.png
  • Anomaly Detection – Graphs the data based on dramatic fluctuations in the frequency (how much that term is used) with which the term exists for a specific time period. The lighter the pink shading the higher the frequency or occurrence of the term regardless of the number of documents.

  • Facets – A dynamic list of values for a specific Metadata field.

  • Facet List – There are three (3) Facet Lists on the Dashboard. The first Facet List is the list of Top Terms within your Dataset. This list will automatically update each time you apply a search or filter to the Dataset to reflect the Top Terms within that subset of data. The other two (2) Facet Lists can be customized based on any available Metadata field ingested into the current Dataset.

    image12.png
  • Facet List Drop-Down Menu – Contains a list of all the available Metadata fields and Extracted Entities. Users can view the top values, based on number of documents, for each Metadata field or Entity in the drop-down menu. By default, the Top Ten values are displayed allowing the user to select one or many values of interest and use those selections to apply a filter to the data by clicking on Add to Search at the bottom of the list of values.

    image13.png
  • Load More – Although Brainspace displays only the top ten values by default for a specific Metadata field, users have the option of loading the next top ten values, based on document volume, by clicking on Load More at the bottom of the value list.

  • Facet Document Count – Brainspace displays the total document count for each unique value of a specific Metadata field in the Facet List.

    image14.png
  • Extracted Entities – Specific people, places and things that have been automatically detected and extracted from the unstructured text of every document regardless of file type. Brainspace extracts fifteen (15) standard Entities from each Dataset. Each Extracted Entity is represented in the Facet List Drop-Down Menu as a Metadata field. Users have the option of filtering on specific Entities and Entity Values. Extracted Entities are either Structured Entities which have repeatable patterns like Credit Card Numbers or Phone Numbers or Unstructured Entities like Geographic Locations, Titles, or Product Names.

    image15.png

Cluster Wheel

An interactive data visualization that groups documents or records into Clusters based on their lexical similarity. Lexically similar documents refer to documents that share the same words or vocabulary. To access the Cluster Wheel data visualization, click on the Cluster Wheel menu option within the Analytics tab. Open the Dataset called English Wiki.

image16.png
  • Cluster – Each wedge on the wheel is called a Cluster. Each Cluster contains a group of documents that are lexically similar or share the same vocabulary. Brainspace displays Cluster Terms or themes within each Cluster.

    image17.png
  • Cluster Terms (Themes) – Each Cluster contains a set of Cluster Terms or themes which best summarize what the documents within the Cluster are about.

  • Cluster Wheel Navigation – Users can Zoom In or Zoom Out at any time while exploring the Cluster Wheel by clicking on either the + or – icon in the lower left-hand corner of the Cluster Wheel. Additionally, users can reset the Cluster Wheel back to its default viewing state by clicking on the clockwise arrow in the lower left-hand corner of the Cluster Wheel.

    image18.png
  • Exploring Cluster Wheel – To explore the Clusters within the Cluster Wheel you can click on a specific Cluster and Brainspace will automatically zoom in on that Cluster and display the Cluster Term Pop-Up window.

  • Cluster Action Menu – This icon provides access to several available actions for the selected Cluster. To access the list of Cluster Actions, click on the Cluster Action Icon in the lower right-hand corner of the selected Cluster.

    • Add Cluster to Search – Click this action to retrieve the documents within the selected Cluster.

    • Add Cluster to Notebook – Click this action to add the documents within the selected Cluster to a specific Notebook (see Notebook section within this Chapter).

    • Add Search Hits to Notebook – Click this action to only add the documents within the selected Cluster that hit on a specific Search (Concept or Keyword) to a specific Notebook.

    • Tag This Cluster – Click this action to apply a Tag to all the documents within the selected Cluster.

    • Tag Only Search Hits – Click this action to only apply a Tag to the documents within the selected Cluster that hit on a specific Search (Concept or Keyword).

      image19.png

Communication Analysis

Communication Analysis – This data visualization provides you with “at a glance” view of the communication patterns between specific people and email domains. You can quickly see which individuals or domains are communicating with one another. In addition, when used in combination with Concept or Keyword Searching, you can see which individuals or domains are communicating about a specific topic or issue.

  • Click on Communications in the upper left-hand corner of the Analytics tab to launch the Communication Analysis visualization.

  • Using a Concept Search example of “Raptor”, you can view the communication activity for individuals having conversations about the topic Raptor.

    image20.png
  • You can click on any Node (circle) to get summary of the communication activity for a specific individual.

    image21.png
  • To review the communications, select a Node and then click on “Add to Query”.

    image22.png

Focus

A Cluster Wheel based on a subset of data within the Dataset: a Focus can be built from any search or filter results allowing you to explore Clusters and Cluster Terms that are specific to a subset of data/documents.

  • To build a Focus you MUST first execute a search or apply a filter to the Dataset. Then simply select the Focus icon in the upper right-hand corner from any screen within the Analyst tab.

  • After selecting the Focus icon the Manage Focus pop-up window will appear. Enter a name for the new Focus in the Name text box. You then have the option of making the Focus Public or Private. Private Focus means that only you can see and interact with the Focus you create. Public Focus means that every user can see and interact with the Focus. By default, the option is set to Private. To change it to Public, click on the slider below the Name field (see screenshot below). You then have the option of including entire Email Threads (Threads), document Families or other Related Documents.

    image23.png
  • Click Save Focus to create the Focus. The Focus will take several minutes to build. You can monitor the status by click on the Focus Drop-Down menu in the upper left-hand corner of the screen. While your Focus is being built it will be listed at the top of the drop-down menu in the Queued section. You will see a Focus Build Status to the right of the Focus name in the drop-down menu. When you first create the Focus, the status will be set to Pending. Once the build process kicks off the status will automatically change to Building. When the Focus finishes building the Focus will appear in the Available section of the Focus Drop-Down menu.

  • To load the Focus, click on the Focus name in the Focus Drop-Down menu after the Focus has finished building and appears in the Available section. You will notice that all of the Data Visualization screen will automatically update to display just the documents that are contained within the Focus. To see the Focus via the Cluster Wheel screen, click on the Cluster Wheel menu option in the upper left-hand corner of the screen. You should see the Cluster Wheel now only includes the documents within your Focus.

Concept Search

The most common first step when managing an investigation, performing data analysis or research, is to perform a Concept Search.

Starting with a phrase, a paragraph, or even an entire document, Brainspace’s Concept Search automatically expands queries to reveal related concepts. These inferences are a result of Brainspace’s understanding of the entire dataset and can quickly introduce key concepts previously unknown to the user.

  • To begin running a Concept Search You will need to navigate to the Analytics tab.

  • You have the ability to run a Concept Search from any of the screens within the Analytics tab. For this example, we are going to want to run a Concept Search from the Cluster Wheel screen. To do this, click on Cluster Wheel in the upper left-hand corner of the screen.

    image24.png
  • To run a Concept Search, enter/paste the term, phrase or paragraph into the Concept Search box located at the top of the screen. The benefit of Concept Searching is that it will allow you to search for specific people, places and things or to search for a loose idea of what you’re looking for. Use the instructions below to execute a Concept Search.

  • To begin, let’s run a Concept Search for the phrase “Pulsars”. To do this enter pulsars into the Concept Search text box and then hit the Enter key on your keyboard.

  • You will immediately notice that the Concept Expansion List pops out below the Concept Search bar. The Concept Expansion List is a series of related concepts (single and multi-word terms). These terms have potentially valuable connections to your Concept Search term(s) based on repeated patterns in the document content.

    image25.png
  • You will also notice that the colorization on the Cluster Wheel has dynamically updated to show you the areas of the wheel that contain documents related to your Concept Search. These areas of the wheel that have a higher concentration of documents relevant to your search will appear in white or light blue shading.

  • Review the terms appearing in the Concept Expansion List to determine if any of these terms may be potentially valuable or interesting based on your initial search.

  • Once you have identified potentially relevant or interesting terms within the Concept Expansion List use the weight selection tool next to the term to make that term required in your search results by selecting the star. Then click “Find content”. This action will further reduce your initial Concept search results.

  • You will notice again that the colorization on the Cluster Wheel has dynamically updated to show the areas of the wheel that have a higher concentration of documents related to now both “pulsars” and “radio pulsars” (white shaded areas of the Cluster Wheel).

    image26.png

Bulk Actions

Users have the option of building a Focus, applying tags to documents (in bulk) or adding documents to a Notebook using the Bulk Action Menu in the upper right-hand corner of every Data Visualization screen within the Analytics tab.

image27.png
  • Create Focus – use the Focus icon to create a Focus on any search or filter results within any screen in the Analytics tab. To create a Focus see Focus section above.

  • Apply Tag – Use the Tag bulk option to apply a Tag/Choice to a group of documents. To apply the Tag to a group of documents click on the Tag icon in the Bulk Action menu and then select the Tag Choice that corresponds with the appropriate Tag Name in the Tag List provided. You have the option to Include Threads, Include Families or Include Related Documents (see Focus section above for a definition of each of these options).

    image28.png
  • Add Documents to Notebook – You can add the results of a search or filter to a new or existing Notebook by clicking on the Notebook icon in the Bulk Actions menu. Then either click New Notebook and create a new Notebook to store the results of your search or filter OR click on an existing Notebook (by name) from the list of existing Notebooks provided below the New Notebook in the list provided.

    image27.png

Saved Searches & Search History

You can create and save searches and access those saved searches on demand. You can also retrieve a search you created but did not save from the Search History feature.

To create a Saved Search, run a Concept (or Keyword or Advanced) search and/or apply any filter to the data and then click on the Floppy Disk icon in the upper right-hand corner of any of the Data Visualization screens within the Analytics tab.

Click on the Save Current Search button in the Saved Search Pop-Up window.

image29.png
  • Enter a name for your Saved Search. You have the option of making the search Private or Public. Click Save Search to save the current search.

    image30.png

Document List & Viewer

  • When you apply a search or filter to your data Brainspace will display the Document List on the right-hand side of the screen.

    image31.png
  • Each card represents a single document. To open a text rendering view of the document, click on the Document Card.

    image32.png

The Document Viewer provides you with a text rendering view of the selected document. In addition, the viewer displays a list of Metadata associated with the document on the left-hand side of the screen. Use your track pad or mouse to scroll down the list of Metadata fields.

image33.png
  • You can also apply a Tag to the document by clicking on the Tags button and then click in Search Tag box. A list of Tags & Choices will be displayed in a Tag List. Select the appropriate Tag/Choice to apply the Tag to this single document. Tags that have already been applied to the document will be displayed just below the Tag List.

    image34.png