Entities  Programming Entities

On this page, you will find information about common entities, including property names and usage information. Click here to show all details. Also, see Recipes and messages. Below is a list of property markers:

calculated Calculated Calculated properties are automatically set by the system, and cannot be changed using update requests.
setonce Set once Set once properties can be set by the user upon entity creation, but cannot subsequently be changed (only read).
filtered Filtered Filtered properties are removed or filtered because they contain confidential information (searches referencing them are also blocked).
Class

WorkspaceInfo

Represents a secure container encapsulating other entities.
Actions: Retrieve, Create, Update, Delete.
Click here to view properties.

  • DatabaseInfoId (Guid calculated):
       ID of the database in which the workspace is stored
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the workspace
  • CreatedBy (String calculated):
       Name of the user who created the workspace
  • Name (String):
    Name of the workspace
  • Description (String):
       Description of the workspace
  • Id (Guid setonce):
       ID of the workspace
  • DateCreated (Date calculated):
       Date the workspace was created

Click here to view usage notes.

  • Many entities have a workspace ID property which controls under which workspace they live.
  • Access to a workspace is controlled by rights, including the right level.
  • For more information about managing rights, click here.
Class

UserInfo

Represents a user account.
Actions: Retrieve, Register, Update.
Click here to view properties.

  • FirstName (String filtered):
       First name associated with the user account
  • LastName (String filtered):
       Last name associated with the user account
  • EmailAddress (String filtered):
       E-mail address associated with the user account
  • Name (String setonce):
    Log in name associated with the user account
  • Password (String filtered):
       Password associated with the user account
  • DisableCompression (Boolean filtered):
    Whether the Web client should try compressing sent XML requests
  • DisableCaching (Boolean filtered):
    Whether the Web client should try caching locally frequently accessed data
  • LicenseKey (String filtered):
    License key associated with the user
  • RegistrationCode (String calculatedfiltered):
    Registration code used to confirm the account via e-mail address registration
  • PasswordResets (Integer calculatedfiltered):
    Number of password resets effected so far
  • RegistrationConfirmations (Integer calculatedfiltered):
    Number of registration confirmations effected so far
  • Disabled (Boolean calculatedfiltered):
    Whether the account has been disabled
  • Id (Guid setonce):
       ID of the user account
  • DateCreated (Date calculatedfiltered):
       Date the user account was created

Click here to view usage notes.

  • You cannot register multiple user accounts sharing the same account name or e-mail address.
  • To protect confidentiality, some properties are filtered when the caller is not the owner of the account (ex: first name, last name, e-mail address, etc.).
  • To ensure security, some properties are always filtered when retrieved, regardless of the caller (ex: registration code, password, etc.).
  • A user's license key can be replaced by another (typically more permissive) one, but never removed.
  • To change account settings, the update request must carry a ticket representing the user associated with the account.
  • If a user was disabled by the system, authentication for this user will fail (this setting cannot be changed).
  • For more information about registering users, click here.
Class

RightInfo

Represents a right granted to a user on a workspace.
Actions: Retrieve, Create, Update, Delete.
Click here to view properties.

  • CreatedBy (String calculated):
       Name of the user who created the right
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace to which the right is granted
  • UserInfoId (Guid setonce):
       ID of the user to which the right is granted
  • RightLevel (Enum):
       Access level granted
  • ApprovalLevel (Enum):
       Current approval level
  • Id (Guid setonce):
       ID of the right
  • DateCreated (Date calculated):
       Date the right was created

Click here to view usage notes.

  • You must have manage permission on a workspace to create (grant) new rights for the workspace.
  • At any given time, at most one right for each user & workspace combination can exist.
  • If no right exists for a given user and workspace, the user has no right whatsoever on the workspace.
  • Workspace owners cannot delete their own rights because this may leave the workspace without any owner.
  • Callers cannot upgrade or downgrade their own right level (another user with manage permission should do this).
  • When a new right is granted, its approval level is set to pending (recipients must update the right and set it to approved to accept, or delete the right to decline).
  • For more information about managing rights, click here.
Class

TableInfo

Represents a table containing either uploaded data, tasks, or task results.
Actions: Retrieve, Update, Delete.
Click here to view properties.

  • CreatedBy (String calculated):
       Name of the user who created the table
  • DatabaseInfoId (Guid calculated):
       ID of the database in which the table is stored
  • WorkspaceInfoId (Guid calculated):
       ID of the workspace encapsulating the table
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the table
  • TableType (Enum calculated):
       Table type.
  • AssociatedTypeName (String calculated):
       Name of an entity whose instances are stored in the table (not applicable to data tables)
  • Ready (Boolean calculated):
       Whether the table is ready because all pending uploaded data chunks have been imported (only applicable to data tables)
  • RowsTruncated (Boolean calculated):
       Whether some imported rows were truncated, due to license restrictions
  • ColumnsTruncated (Boolean calculated):
       Whether some imported fields (columns) were truncated, due to license restrictions
  • Name (String calculated):
    Name of the table
  • Description (String):
       Description of the table
  • Id (Guid calculated):
       ID of the table
  • DateCreated (Date calculated):
       Date the table was created

Click here to view usage notes.

  • Tables are automatically created when uploading new data, and cannot be created using normal create message requests.
  • Data tables contain data sets uploaded by users, while other types of tables may contain tasks, task results, etc.
  • Tables names are SQL names which are automatically generated and cannot be set (however the description can be set).
  • For more information about uploading data, click here.
  • For more information about executing tasks, click here.
Class

FieldInfo

Represents a field (column) within a data table.
Actions: Retrieve.
Click here to view properties.

  • TableInfoId (Guid calculated):
       ID of the data table associated with the field
  • Name (String calculated):
    Name of the field
  • Description (String calculated):
    Description of the field
  • FieldType (Enum calculated):
    Field type
  • FieldPurpose (Enum calculated):
    Field purpose
  • IsSystem (Boolean calculated):
    Whether the field was not present in uploaded data, but added for tracking purposes (ex: system field to identify each row using a unique ID)
  • TotalCount (Integer calculated):
    Total number of rows
  • NotNullCount (Integer calculated):
    Total number of rows for which the field is not null / missing
  • DistinctCount (Integer calculated):
    Total number of rows for which the field has distinct values
  • Avg (Double calculated):
    Average value for the field (only applicable to numeric fields)
  • Std (Double calculated):
    Standard deviation for the field (only applicable to numeric fields)
  • Min (Double calculated):
    Min value for the field (only applicable to numeric fields)
  • Max (Double calculated):
    Max value for the field (only applicable to numeric fields)
  • IsIndexed (Boolean calculated):
    Whether the system decided to index the field
  • FieldOrder (Integer calculated):
    Order in which the field appeared when data was imported (ex: first column = order 0)
  • DistributionStatXml (String calculated):
    XML specifying the distribution of data (ex: top discrete values, distribution of numeric values)
  • Id (Guid calculated):
       ID of the field

Click here to view usage notes.

  • Fields are automatically created when importing new data, and cannot be created using normal create message requests.
  • Field names are SQL names which are automatically generated and cannot be set.
  • Fields cannot be updated or deleted directly (delete the associated table).
  • For more information about uploading data, click here.
Class

ChunkInfo

Represents a chunk of uploaded data used to create a new data table.
Actions: Retrieve, Create, Delete.
Click here to view properties.

  • CreatedBy (String calculated):
       Name of the user who created the data chunk
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace encapsulating the data chunk
  • TableInfoId (Guid setonce):
       ID of the data table to create when importing all data chunks
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the data chunk
  • FileName (String setonce):
       Name of the file from which the data chunk was read
  • Content (String setonce):
       Base64-encoded bytes from a CSV file
  • ChunkOrder (String setonce):
       Order in which the data chunk appeared in the data file (ex: first chunk = order 0)
  • HasFieldHeader (Boolean setonce):
       Whether the first line in the data file specifies field names (vs. contains the first row of data)
  • Id (Guid setonce):
       ID of the data chunk
  • DateCreated (Date calculated):
       Date the data chunk was created

Click here to view usage notes.

  • Do not forget to increment the chunk order as you read and upload more data chunks from a file.
  • Ensure you specify the same data table ID while uploading data chunks from the same file.
  • Ensure the specified data table ID does not already exist at upload time (it will be created later on by executing a data upload task).
  • Data chunks cannot be updated but can be deleted.
  • Data chunks which are created but not committed to a data table count against your license quota (commit or delete them).
  • For more information about uploading data, click here.
Class

ImageInfo

Represents an image to display in the gallery.
Actions: retrieve, create, update, delete
Click here to view properties.

  • CreatedBy (String calculated):
       Name of the user who created the image
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace encapsulating the image
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the image
  • ThumbnailContent (String calculated):
       Base64-encoded image thumbnail content (in png format)
  • ImageContent (String setonce):
       Base64-encoded image content (in png format)
  • XamlContent (String setonce):
       Base64-encoded image content (in XAML format)
  • Keywords (String):
       Column-separated list of keywords associated with the image
  • Size (Long calculated):
       Size of the image (in bytes)
  • Width (Integer calculated):
       Width of the image (in pixels)
  • Height (Integer calculated):
       Height of the image (in pixels)
  • ThumbnailWidth (Integer calculated):
       Width of the thumbnail image (in pixels)
  • ThumbnailHeight (Integer calculated):
       Height of the thumbnail image (in pixels)
  • Rating (Double):
       Average rating for the image (from 0 to 5), or new rating (see usage notes)
  • RatingCount (Integer):
       Total number of ratings for the image
  • Name (String):
    Name of the image
  • Description (String):
       Description of the image
  • Id (Guid setonce):
       ID of the image
  • DateCreated (Date calculated):
       Date the image was created

Click here to view usage notes.

  • To create an image, you can specify XAML or png content, but not both.
  • If you specify XAML content, equivalent png content will be automatically generated and set.
  • Thumbnail content and image size information are automatically generated and set.
  • The rating specifies the average rating.
  • To rate the image, set the rating property and update the image: the average rating will be adjusted and updated.
Class

CommentInfo

Represents a comment associated with an image or task results.
Actions: Retrieve, Create, Update, Delete.
Click here to view properties.

  • CreatedBy (String calculated):
       Name of the user who created the comment
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace encapsulating the comment
  • OwnerId (Guid setonce):
       ID of the entity with which the comment is associated (ex: image ID, task result ID)
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the comment
  • Tag (String):
       Tag indicating which particular item of a task result the comment is associated with
  • Comment (String):
       Textual comment
  • Color (Integer):
       Color used to show the comment in the Web client
  • Id (Guid setonce):
       ID of the comment
  • DateCreated (Date calculated):
       Date the comment was created

Click here to view usage notes.

  • Comments are typically associated with images or task results.
  • For tasks, a tag can be used to indicate which particular item of a task result the comment should be associated with.
  • For example a tag may be used to indicate that a comment should be associated with a particular area of a tree map result.
  • To prevent abuse, comments can only be updated by their owners.
Class

SettingInfo

Represents a setting used to persist any user preference.
Actions: Retrieve, Create, Update, Delete.
Click here to view properties.

  • CreatedBy (String calculated):
       Name of the user who created the setting
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace encapsulating the setting
  • OwnerId (Guid setonce):
       ID of the entity with which the setting is associated
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the setting
  • Name (String):
       Name of the setting
  • Value (String):
       Value of the setting
  • Id (Guid setonce):
       ID of the setting
  • DateCreated (Date calculated):
       Date the setting was created

Click here to view usage notes.

  • Settings can be used to persist any user preference using a name / value pair.
  • For example, the Web client persists information about how users arranged views using settings.
  • To prevent abuse, settings can only be updated by their owners.
  • To prevent abuse, a limited number of settings can be created per workspace and per user.
Class

LogInfo

Represents a security log event.
Actions: Retrieve.
Click here to view properties.

  • WorkspaceInfoId (Guid calculated):
       ID of the workspace encapsulating the log event
  • UserName (String calculated):
       Name of the user who caused an event to be logged
  • Action (String calculated):
       Name of the action performed
  • Target (String calculated):
       Name of the entity on which the action was performed
  • Details (String calculated):
       Additional details about the action performed
  • Id (Guid calculated):
       ID of the log event
  • DateCreated (Date calculated):
       Date the log event was created

Click here to view usage notes.

  • Log events are automatically created when actions are performed on a workspace.
  • Only users with manage permission on a workspace can retrieve corresponding log events.
  • Log events cannot be deleted or updated.
Class

OrgLicenseInfo

Represents an organization license.
Actions: Retrieve, Create.
Click here to view properties.

  • LicenseKey (String calculated):
       License key
  • Id (Guid setonce):
       ID of the org license
  • DateCreated (Date calculated):
       Date the org license was created

Click here to view usage notes.

  • Organization licenses can be added to unlock additional capabilities, but not updated or deleted.
  • For more information about licensing, click here.
Class

RootDataUploadTaskInfo

Represents a data import task (imports data following data chunk upload).
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • LineSeparator (String setonce):
       Separator to use to parse uploaded CSV data (applies only to CSV data, defaults to "," if unspecified)
  • CultureName (String setonce):
       Culture to use to parse uploaded CSV data (applies only to CSV data, defaults to the neutral culture if unspecified)
  • FileType (String setonce):
       File type (ex: ".csv", ".xls")
  • Context (String setonce):
       Upload context (for Excel files: name of the data sheet to upload, required only when multiple ones exist)
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and into which data should be imported
  • TargetTableInfoId (Guid setonce):
       ID of the (not yet created) data table for which data chunks have already been uploaded
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider (not applicable)
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data (not applicable)
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task is designed to import uploaded data, and does not produce any type of task result.
  • You can specify how uploaded content is formatted using properties LineSeparator and CultureName.
  • For example, US CSV files may contain dates in MM/DD/YY format, but a different format may be used by other locales.
  • For Excel files, you can specify the data sheet to upload using property Context (required only if the Excel file contains multiple data sheets).
  • Finally, the file type should be specified, or will be assumed to be ".csv".
  • For more information about uploading data, click here.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootDataTransformTaskInfo

Represents a data transformation task (transforms an existing table into a new one).
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • TargetTableDescription (String setonce):
       Description to assign to the post-transform data table to create
  • SourceTableInfoId (Guid setonce):
       ID of the existing data table to transform
  • TransformSequenceXml (String setonce):
       XML specifying a sequence of transformation steps to execute
  • TransformEditSequenceXml (String setonce):
       XML used by the Web client to keep track of editing state
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and into which a transformed data table should be created
  • TargetTableInfoId (Guid setonce):
       ID of the post-transform data table to create
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider (not applicable)
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data (not applicable)
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task is designed to transform an existing data table so as to produce a new one.
  • You can simply copy an existing data set by setting property TransformSequenceXml to an empty string.
  • Or you can specify a sequence of transformation steps to execute by serializing a TransformSequence object.
  • Different types of available transformation steps are listed on this page.
  • For example, you can create, rename, delete, or convert fields.
  • For example, you can filter or sample rows.
  • For example, you can rank rows, scramble fields, or set fields to calculated formulas.
  • For more information about transforming data, click here.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootPivotTaskInfo

Represents a super pivot task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data (not applicable)
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task is used for data visualization, and does not produce any type of task result.
  • For more information about uploading data, click here.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootTimeSeriesTaskInfo

Represents a time series forecasting task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • TargetFieldInfoId (Guid setonce):
       ID of the data field containing numeric values to forecast
  • SourceFieldInfoId (Guid setonce):
       ID of the data field containing corresponding time series dates
  • Predictions (Integer setonce):
       Number of values (at future time steps) to forecast
  • Simulations (Integer setonce):
       Number of monte-carlo simulations executed so as to estimate the stability of the forecasting model
  • TrainingTime (Integer setonce):
       Maximum execution time allowed (in milliseconds) when training the neural network on the time series
  • TrainingRounds (Integer setonce):
       Maximum number of training rounds allowed when training the neural network on the time series
  • LearningRate (Double setonce):
       Learning rate used by the neural network (typical values range from 0.1 to 0.9)
  • Error (Double calculated):
       Specifies the average error obtained after training the neural network on the time series (measures how closely the model fits the time series)
  • LagCount (Integer setonce):
       Number of past time intervals to consider when training the neural network on the time series (0 = autodetect)
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider (not applicable)
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data (not applicable)
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of TimeSeriesResultInfo records.
  • A neural network is trained on time series data to identify trends & patterns, and use this knowledge to forecast future values.
  • Once training is complete, the difference (error) between known past values and values predicted by the forecasting model is measured.
  • The distribution of errors is used to perform monte-carlo simulations, which help visualize the stability of the forecasting model.
  • For example, if the neural network was unable to build a forecasting model which closely fits the time series, errors will be high.
  • Accordingly, higher variations from predicted values should be expected.
  • The neural network learns how the previous N values (the lag) influence the following one.
  • The neural network is able to automatically detect the general periodicity of the time series.
  • The neural network automatically compensates for missing values or irregular time intervals.
  • The neural network stops training when lack of convergence is detected.
  • The neural network keeps track of the best model found during training.
  • The neural network uses a configuration designed to avoid overfitting the data and preserve generalization power.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootTreeMapTaskInfo

Represents a tree map task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data (not applicable)
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task is used for data visualization, and does not produce any type of task result.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootCorrelationTaskInfo

Represents a correlation analysis task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of CorrelationResultInfo records.
  • A measure of the correlation between each pair of numeric variable is calculated.
  • Since correlation is valid only for numeric values, non-numeric fields are ignored.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootOutlierDetectionTaskInfo

Represents an outlier detection analysis task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • OutlierCount (Integer setonce):
       Number of top outliers to find (keep this value low for performance reasons)
  • OutlierNeighbors (Integer setonce):
       Number of neighbors to consider to find outliers (keep this value low for performance reasons)
  • OutlierCandidateBatchSize (Integer setonce):
       Number of data rows loaded together before being compared with other data rows
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of OutlierDetectionResultInfo records.
  • Data rows are loaded in batches, and compared to other data rows.
  • Data rows whose average distance to their closest N neighbors is the greatest are considered outliers.
  • Callers must specify the batch size, how many outliers to find, and how many neighbors to consider.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootAssociationTaskInfo

Represents an association rule analysis task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • MaxNumericBuckets (Integer setonce):
       Number of buckets used to discretize numeric values
  • MaxDiscreteBuckets (Integer setonce):
       Number of buckets used to process discrete values
  • TargetDimension (Integer setonce):
       Size of association rules to find (typical values range from 2 to 4)
  • MinDensityThreshold (Double setonce):
       Minimum density threshold required to keep a discovered association between fields
  • MaxDensityHypercubes (Integer setonce):
       Maximum number of discovered associations to keep during each pass
  • ShrinkNumericDensityIntervals (Boolean setonce):
       Whether the description of association rules should be modified to fit processed data
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of AssociationResultInfo records.
  • Numeric variables are discretized by taken into consideration N equal-sized numeric buckets (ex: age field: 0 to 20, 20 to 40, etc.).
  • Discrete variables are constrained by taken into consideration only the most frequent N discrete values.
  • Then, combinations of such variables (called hypercubes) are generated, and occurences of each combination counted.
  • Hypercubes are combined with other hypercubes, generating larger combinations of variables.
  • Callers can specify the maximum number of hypercubes so as to prevent excessive memory usage.
  • Callers can specify the minimum number of instances (called density) required to keep a hypercube.
  • The density is simply the percentage of data rows matching the hypercube (ex: 0.01 = 1% of data rows analyzed).
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootNumericDecisionTreeTaskInfo

Represents a numeric decision tree task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • MaxEquiDistributionProbability (Double setonce):
       Required min probability of having different child distributions as a result of a split
  • TargetFieldInfoId (Guid setonce):
       ID of the data field containing numeric values and for which to build a decision tree
  • MaxDepth (Integer setonce):
       Maximum depth allowed for the decision tree
  • MaxSplit (Integer setonce):
       Maximum number of child branches allowed when splitting on a discrete value
  • MinDensityThreshold (Double setonce):
       Minimum density threshold required to split further
  • CurrentDepth (Integer setonce):
       Depth at which the task is currently processing data
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of DecisionTreeResultInfo records.
  • This task can be used to build decision trees for a numeric target field.
  • At each level, a split using different fields and values is considered, and the target field's distribution is observed.
  • The algorithm selects the split having the most impact on the target field's distribution.
  • Statistical tests are used to determine how different distributions would be in each child branch, if a split was performed.
  • If distributions appear quite similar between child branches, then no split is performed.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootDiscreteDecisionTreeTaskInfo

Represents a discrete decision tree task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • MinGainRatioImprovement (Double setonce):
       Min gain ratio improvement required to create a split
  • MinEntropy (Guid setonce):
       Min entropy required to split further
  • TargetFieldInfoId (Guid setonce):
       ID of the data field containing numeric values and for which to build a decision tree
  • MaxDepth (Integer setonce):
       Maximum depth allowed for the decision tree
  • MaxSplit (Integer setonce):
       Maximum number of child branches when splitting on a discrete value
  • MinDensityThreshold (Double setonce):
       Minimum density threshold required to split further
  • CurrentDepth (Integer setonce):
       Depth at which the task is currently processing data
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of DecisionTreeResultInfo records.
  • This task can be used to build decision trees for a discrete target field.
  • At each level, a split using different fields and values is considered, and the target field's distribution is observed.
  • The algorithm selects the split having the most impact on the target field's distribution.
  • Statistical tests are used to determine how different distributions would be in each child branch, if a split was performed.
  • If distributions appear quite similar between child branches, then no split is performed.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootClusteringTaskInfo

Represents a cluster identification analysis task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • AutoCluster (Enum setonce):
       Auto-cluster mode to use
  • ClusterCount (Integer setonce):
       Specific number of clusters to identify (used only when manual mode is selected)
  • Outlier ratio (Double setonce):
       Value from 0 to 1 indicating by how much cluster models should be simplified so as to filter out noise & outliers
  • MaxClusterCount (Integer setonce):
       Maximum number of clusters to identify
  • ClustersFound (Integer calculated):
       Number of clusters identified
  • PointsClustered (Integer calculated):
       Number of data rows found to belong to one of the identified clusters
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of ClusteringResultInfo and ClusteringSummaryResultInfo records.
  • In manual mode, an exact number of clusters can be specified, for example 3 clusters to divide data rows into 3 groups.
  • In automatic mode, several clustering models are generated, and a best fit is selected.
  • To prevent automatic mode from generating models with an excessive number of clusters, a maximum cluster count can be specified.
  • To prevent noise or outliers from affecting clustering quality, an outlier ratio can also be specified.
  • For example, a 0.25 value means that cluster models will be simplified by 25% before data rows are assigned to each cluster.
  • Points that do not fit one of the simplified clusters will be reported as outliers.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

RootSomMapTaskInfo

Represents a similarity map task.
Actions: Retrieve, Create, Update, Delete, Start.
Click here to view properties.

  • Width (Integer setonce):
       Width of the map (in number of cells)
  • Height (Integer setonce):
       Height of the map (in number of cells)
  • Iterations (Integer setonce):
       Number of training iterations to perform
  • MaxDiscreteValueInit (Integer setonce):
       Maximum number of discrete elements to consider when initializing the map
  • ReinforcementCutoff (Double setonce):
       Reinforcement cutoff for propagating to other cells
  • WorkspaceInfoId (Guid setonce):
       ID of the workspace under which the task lives and containing data to process
  • TargetTableInfoId (Guid setonce):
       ID of the data table containing data to process
  • OwningUserInfoId (Guid calculated):
       ID of the user owning the task
  • Description (String):
       Description of the task
  • CreatedBy (String calculated):
       Name of the user who created the task
  • SampleSize (Integer setonce):
       Number of random samples to consider
  • FieldInfoFilterXml (String setonce):
       XML specfifying which fields to consider when processing data
  • SendNotification (Boolean):
       Whether an e-mail should be sent upon task completion or failure
  • RunState (Enum calculated):
       Current task execution state
  • RunStatus (String calculated):
       Additional information about the task execution state (ex: progress, error, etc.)
  • ErrorCode (Enum calculated):
       Current task execution error code
  • Generation (Integer calculated):
       Generation of the task (normally 0, but 1 or higher for failed & restarted tasks)
  • Id (Guid setonce):
       ID of the task
  • DateCreated (Date calculated):
       Date the task was created

Click here to view usage notes.

  • This task generates task results in the form of SomMapResultInfo and SomMapDistanceResultInfo records.
  • Data rows are projected to a 2D map made of cells, using a topology preserving algorithm.
  • This allows similar data rows to be mapped to identical or neighboring cells.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

TimeSeriesResultInfo

Represents a time series forecasting task result.
Actions: Retrieve.
Click here to view properties.

  • Date (Date calculated):
       Date of the forecast
  • Value (Double calculated):
       Value of the forecast
  • Min100 (Double calculated):
       Min value when taking into consideration 100% of monte-carlo deviations
  • Max100 (Double calculated):
       Max value when taking into consideration 100% of monte-carlo deviations
  • Min95 (Double calculated):
       Min value when taking into consideration the most central 95% monte-carlo deviations
  • Max95 (Double calculated):
       Max value when taking into consideration the most central 95% monte-carlo deviations
  • Min85 (Double calculated):
       Min value when taking into consideration the most central 85% monte-carlo deviations
  • Max85 (Double calculated):
       Max value when taking into consideration the most central 85% monte-carlo deviations
  • Min75 (Double calculated):
       Min value when taking into consideration the most central 75% monte-carlo deviations
  • Max75 (Double calculated):
       Max value when taking into consideration the most central 75% monte-carlo deviations
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootTimeSeriesTaskInfo tasks.
  • Each result represents a single forecast, and includes an estimate of possible variations.
  • For more information about executing tasks, click here.
Class

CorrelationResultInfo

Represents a correlation analysis task result.
Actions: Retrieve.
Click here to view properties.

  • XFieldId (Guid calculated):
       First field correlation was calculated for
  • YFieldId (Double calculated):
       Second field correlation was calculated for
  • Correlation (Double calculated):
       Estimate of the correlation
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootCorrelationTaskInfo tasks.
  • Each result specifies the correlation between a pair of numeric fields.
  • For more information about executing tasks, click here.
Class

OutlierDetectionResultInfo

Represents an outlier detection task result.
Actions: Retrieve.
Click here to view properties.

  • AverageDistance (Double calculated):
       Average distance to neighbors of the outlier
  • Random (Integer calculated):
       Random value designating a data row which is an outlier
  • NeighborsXml (String calculated):
       XML specifying the random values of neighbors and the distance to these neighbors
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootOutlierDetectionTaskInfo tasks.
  • Each result corresponds to a single outlier, and specifies neighbors and distance information.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

AssociationResultInfo

Represents an association rule task result.
Actions: Retrieve.
Click here to view properties.

  • Expected (Integer calculated):
       Count of data rows expected to satisfy the association's preconditions assuming independence between variables
  • Actual (Integer calculated):
       Count of data rows found to satisfy the association's preconditions
  • Strength (Double calculated):
       Value representing the strength of the association
  • RuleXml (String calculated):
       XML specifying association preconditions (priors)
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootAssociationTaskInfo tasks.
  • Each result corresponds to a single association, and specifies preconditions for the association.
  • For more information about executing tasks, click here.
Class

DecisionTreeResultInfo

Represents a decision tree task result (i.e. a split).
Actions: Retrieve.
Click here to view properties.

  • ParentDecisionTreeResultInfoId (Guid calculated):
       ID of the parent decision tree task result (empty GUID if root split)
  • SplitFieldInfoId (Guid calculated):
       ID of the field which was used for the current split
  • SplitFieldRestrictionXml (String calculated):
       XML specifying the split (ex: "greater than X")
  • SplitFieldValue (String calculated):
       String representing the split value (ex: if split on "greater than X" -> X)
  • SplitFieldCondition (Condition calculatedobject):
       Condition representing the split operator (ex: "greater than X" -> greater)
  • DistributionStatXml (String calculated):
       XML specifying the target field's distribution before splitting
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootNumericDecisionTreeTaskInfo & RootDiscreteDecisionTreeTaskInfo tasks.
  • Each result corresponds to a single split, and specifies how the split was effected, as well as information about the target field's distribution.
  • For more information about executing tasks, click here.
Class

ClusteringResultInfo

Represents a cluster detection task result used to report cluster memberships.
Actions: Retrieve.
Click here to view properties.

  • Cluster (Integer calculated):
       Integer value representing a cluster
  • Random (Guid calculated):
       Random value designating a data row which is a member of the cluster
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootClusteringTaskInfo tasks.
  • Each result specifies that a single data row is a member of a cluster.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

ClusteringSummaryResultInfo

Represents a cluster detection task result used to report cluster profiles.
Actions: Retrieve.
Click here to view properties.

  • Cluster (Integer calculated):
       Integer value representing a cluster
  • Count (Integer calculated):
       Number of data rows found to belong to the cluster
  • Outliers (Integer calculated):
       Whether the cluster represents outliers or noise records
  • MultiDistributionStatXml (String calculated):
       XML specifying the distribution taken by all data fields for all members of the cluster
  • ComplementMultiDistributionStatXml (String calculated):
       XML specifying the distribution taken by all data fields for all non-members of the cluster
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootClusteringTaskInfo tasks.
  • Each result specifies profile information for a single cluster.
  • For more information about executing tasks, click here.
Class

SomMapResultInfo

Represents a similarity map task result used to report cell memberships.
Actions: Retrieve.
Click here to view properties.

  • Random (Integer calculated):
       Random value designating a data row which was mapped to the cell
  • X (Integer calculated):
       X-coordinate of a cell on the similarity map
  • Y (String calculated):
       Y-coordinate of a cell on the similarity map
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootSomMapTaskInfo tasks.
  • Each result specifies that a single data row was mapped to a cell.
  • For more information about executing tasks, click here.
  • For more information about sampling & identification, click here.
Class

SomMapDistanceResultInfo

Represents a similarity map task result used to report cell profiles.
Actions: Retrieve.
Click here to view properties.

  • X (Integer calculated):
       X-coordinate of a cell on the similarity map
  • Y (String calculated):
       Y-coordinate of a cell on the similarity map
  • DistanceRight (Double calculated):
       Distance to the cell located immediately to the right of the current cell
  • DistanceBelow (Double calculated):
       Distance to the cell located immediately below the current cell
  • Count (Integer calculated):
       Number of data rows found to be mapped to the cell
  • MultiDistributionStatXml (String calculated):
       XML specifying the distribution taken by all data fields for all data rows mapped to the cell
  • Id (Guid calculated):
       ID of the task result
  • OwnerId (Guid calculated):
       ID of the task owning the task result
  • Generation (Integer calculated):
       Generation of the task owning the task result (normally 0, but 1 or higher for failed & restarted tasks)

Click here to view usage notes.

  • This type of task result is generated by RootSomMapTaskInfo tasks.
  • Each result specifies profile information for a single cell.
  • For more information about executing tasks, click here.

The following objects and enums are used to support entities:

Object

RunState

Enum specifying the state a task is in.
Click here to view properties.

  • None = 0: the task has not been started
  • Scheduled = 1: the task has been started, but execution has not begun
  • Started = 2: the task has been started, and is being processed
  • Completed = 3: the task completed successfully
  • Failed = 4: the task failed during execution

Click here to view usage notes.

  • Tasks which have been created but not yet started have a RunState of None.
  • Tasks which have been started but for which execution has not begun have a RunState of Scheduled.
  • Tasks which have been started and are being processed have a RunState of Started (check the RunStatus property for progress).
  • Tasks which completed successfully have a RunState of Completed.
  • Tasks which failed during execution have a RunState of Failed (check the RunStatus and ErrorCode properties for details).
  • For more information about executing tasks, click here.
Object

RightLevel

Enum specifying the right level granted to a user on a workspace.
Click here to view properties.

  • Read = 0: the right confers read access to a workspace
  • Write = 1: the right confers write access to a workspace
  • Manage = 2: the right confers manage access to a workspace

Click here to view usage notes.

  • For more information about managing rights, click here.
Object

ApprovalLevel

Enum specifying the approval level for a given right.
Click here to view properties.

  • PendingApproval = 0: the right has been granted, but has not been approved by the recipient
  • Approved = 1: the right has been granted, and has been approved by the recipient

Click here to view usage notes.

  • For more information about managing rights, click here.
Object

TableType

Enum specifying the type of data stored by a table.
Click here to view properties.

  • Data = 0: the table stores uploaded data
  • RootAlgorithm = 1: the table stores a root task
  • ChildAlgorithm = 2: the table stores a child task spawned by a root task
  • AlgorithmResult = 3: the table stores task results

Click here to view usage notes.

  • For more information about uploading data, click here.
  • For more information about executing tasks, click here.
Object

FieldType

Enum specifying the type of data stored by a data field.
Click here to view properties.

  • UniqueIdentifier = 0: the field stores GUIDs
  • Text = 1: the field stores text
  • Bit = 2: the field stores booleans
  • Int = 3: the field stores integers
  • BigInt = 4: the field stores long integers
  • Double = 5: the field stores double-precision numeric values
  • Date = 6: the field stores dates

Click here to view usage notes.

  • For more information about uploading data, click here.
  • For more information about sampling & identification, click here.
Object

FieldPurpose

Enum specifying the type of data stored by a data field.
Click here to view properties.

  • Unknown = 0: the field stores unknown data
  • Date = 1: the field stores temporal data
  • Rating = 2: the field stores rating data
  • Money = 3: the field stores monetary data
  • ScientificMeasure = 4: the field stores scientific measurement data
  • Location = 5: the field stores geographic data
  • Count = 6: the field stores count data
  • Classification = 7: the field stores classification data
  • Description = 8: the field stores description data
  • Naming = 9: the field stores naming data
  • Duration = 10: the field stores duration data
  • Proportion = 11: the field stores proportion data
  • RandomIdentity = 12: the field stores an identity used for random selection
  • HashSelection = 13: the field stores a hash value used for record indexing or selection

Click here to view usage notes.

  • For more information about uploading data, click here.
  • For more information about sampling & identification, click here.
Object

AutoCluster

Enum specifying a clustering mode.
Click here to view properties.

  • Manual = 0: identify an exact number of clusters.
  • Best = 1: automatically identify the best number of clusters.
  • Low = 2: automatically identify the best number of clusters, but favor models with a lower number of clusters.
  • High = 3: automatically identify the best number of clusters, but favor models with a higher number of clusters.

Click here to view usage notes.

  • For more information about executing tasks, click here.

The following objects are used to support data transformation tasks:

Object

TransformSequence

Represents a sequence of transformation steps to execute.
Click here to view properties.

  • Transforms (Transform[] entityarray ):
       Sequence of transformation steps to execute

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

CreateTransform

Represents a transformation step which creates a field.
Click here to view properties.

  • FieldInfoId (Guid):
       ID to assign to the field to create
  • FieldDescription (String):
       Description of the field to create
  • FieldType (Enum):
       Type of field to create (Enum)
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

RenameTransform

Represents a transformation step which renames a field.
Click here to view properties.

  • FieldInfoId (Guid):
       ID of the field to rename
  • FieldDescription (String):
       New description of the field to rename
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

ConvertTransform

Represents a transformation step which converts a field to a new type.
Click here to view properties.

  • FieldInfoId (Guid):
       ID of the field to convert
  • FieldType (Enum):
       Field type to convert to
  • FieldDefaultValue (Value object):
       Field default value if conversion fails
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • Set the default value based on the new field type.
  • For example, for a date field, specify a date value, or null (for missing).
  • For more information about transforming data, click here.
Object

DeleteTransform

Represents a transformation step which deletes a field.
Click here to view properties.

  • FieldInfoId (Guid):
       ID of the field to delete
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

SampleTransform

Represents a transformation step which randomly samples data.
Click here to view properties.

  • SampleMode (SampleMode object):
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • Set the sample mode to a CountSample or ProportionSample object.
  • For more information about transforming data, click here.
Object

FilterTransform

Represents a transformation step which filters rows of data.
Click here to view properties.

  • DeleteRows (bool):
       Whether rows matching the condition restriction should be kept or skipped
  • ConditionRestriction (SelectConditionRestriction object):
       Filtering condition restriction specifying which rows to filter or skip
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

RankTransform

Represents a transformation step which ranks rows and then sets a field to the rank.
Click here to view properties.

  • FieldInfoId (Guid):
       ID of the field to receive the rank
  • WithTies (bool):
       Whether ties are allowed when ranking rows
  • ConditionRestriction (SelectConditionRestriction object):
       Filtering and ordering condition restriction specifying which rows to rank, and in which order to rank them
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

ScrambleTransform

Represents a transformation step which replaces a field by a scrambled text equivalent.
Click here to view properties.

  • FieldInfoId (Guid):
       ID of the text field to receive scrambled values
  • ScrambledFieldInfoId (Guid):
       ID of the field to scramble
  • Prefix (Guid):
       Optional prefix to use when scrambling data
  • ConditionRestriction (SelectConditionRestriction object):
       Filtering condition restriction specifying which rows to apply scrambling to
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • If two original values are identical, they scrambled equivalents will be the same as well.
  • This makes it possible to know that two values are identical, but without knowing their original value.
  • For more information about transforming data, click here.
Object

SetTransform

Represents a transformation step which sets a field to a fixed or dynamic expression.
Click here to view properties.

  • FieldInfoId (Guid):
       ID of the field to set
  • Expression (Text):
       Fixed or dynamic expression
  • FieldDefaultValue (Value object):
       Field default value if computing or setting the calculated value fails
  • ConditionRestriction (SelectConditionRestriction object):
       Filtering condition restriction specifying which rows to set a field for
  • Description (String):
       Optional description for the transformation step

Click here to view usage notes.

  • Set the default value based on the field type.
  • For example, for a date field, specify a date value, or null (for missing).
  • Example fixed numeric expression: 0.13.
  • Example dynamic numeric expression: [field1_name] + log([field2_name] + 0.13).
  • Example fixed text expression: "X".
  • Example dynamic text expression: [field1_name] + sub([field2_name], 0, 2) + "X".
  • Example fixed boolean expression: false.
  • Example dynamic boolean expression: (10 < [field1_name]) || [field2_name].
  • Example fixed date expression: #12/31/2010#.
  • Example dynamic date expression: addday([field1_name], day([field1_name])).
  • Logical functions: (, ), <, <=, >, >=, =, ==, !=, <>, !, not, ||, or, &&, and
  • Text functions: +, len(x), sub(x,pos,len), repl(x,old,new), trim(x), lower(x), upper(x), indexof(x,v), startsw(x,v), endsw(x,v), contains(x,v)
  • Numeric functions: +, -, *, /, ^, %, min(x,y), max(x,y), abs(x), sign(x), floor(x), ceil(x), round(x,prec), exp(x), log(x), rand()
  • Date functions: msec(x), sec(x), day(x), mth(x), yr(x), addmsec(x,v), addsec(x,v), addmin(x,v), addday(x,v), addmth(x,v), addyr(x,v)
  • For more information about transforming data, click here.
Object

CountSampleMode

Represents a sampling mode where a specified number of rows are sampled at random.
Click here to view properties.

  • Count (Integer):
       Number of rows to select

Click here to view usage notes.

  • For more information about transforming data, click here.
Object

ProportionSampleMode

Represents a sampling mode where a proportion of rows are sampled at random.
Click here to view properties.

  • Proportion (Double):
       Proportion of rows to select (from 0 to 1)

Click here to view usage notes.

  • For more information about transforming data, click here.