Data Type

Can be one of the following types:

Overview

The 'Extractor' property allows you to assign a single Value Extractor to this Data Type. This is ideal for simple extraction scenarios where only one extraction method is needed.

Execution Order and Combined Extractors

If 'Extractor' is set, and you also configure 'Extractors' (referenced extractors) or add child extractors. All configured extractors will be executed in the following order:

1. The local extractor (this property)
2. Any direct child extractors
3. Any referenced extractors from the 'Extractors' property

The results from all extractors are merged according to the selected 'Collation' method.

This allows you to combine multiple extraction strategies within a single Data Type, and to control the order in which their results are considered and merged.

Configuration

• Select the desired Value Extractor type from the drop-down menu.
• Configure its properties (such as pattern, region, or barcode type) as needed for your extraction scenario.
• If you later need to use multiple extractors, you may set this property, add child extractors, and/or reference additional extractors as needed.

Advanced Usage

• The local extractor is best for straightforward cases, such as extracting a single value with a regular expression, barcode reader, or zonal region.
• If you need to combine multiple extraction strategies (for example, to capture values in different formats or locations), use the 'Extractors' property or add child extractors.
• The local extractor can be used in conjunction with input, exclusion, or subtraction extractors for additional filtering and refinement.

Best Practices

• Use the local extractor for the most common or reliable extraction method.
• For maintainability, prefer referenced or child extractors when logic becomes complex or needs to be reused.

Overview

The 'Extractors' property allows you to reference additional Extractor Nodes (such as Value Readers, Field Classes, or other Data Types) to participate in extraction.

Execution Order

• Referenced extractors are executed after the local extractor and any direct child extractors.
• The order of execution is determined by the order of items in this list (top to bottom).
• All results are merged according to the selected 'Collation' method.

Configuration

• Use the collection editor to select one or more extractors from the available nodes in your project or local resources.
• Change the order using the up/down arrows to control execution sequence.
• You can reference extractors from other Data Types, Value Readers, or Field Classes to modularize and share extraction logic.

Advanced Usage

• Use referenced extractors to combine multiple extraction strategies, such as handling different value formats, languages, or document layouts.
• This property is especially useful for complex data elements that require fallback logic or aggregation from multiple sources.
• Referenced extractors can be reused across multiple Data Types, promoting consistency and reducing duplication.

Best Practices

• Use referenced extractors to modularize and share extraction logic.
• Keep the list concise and well-ordered to ensure predictable results.
• Document the purpose of each referenced extractor for future maintainers.

Can be one of the following types:

Overview

The 'Collation' property determines the method used to combine results from all configured extractors. The selected Collation Provider controls how individual matches are grouped, ordered, or structured in the output.

Collation Methods

• Individual: Returns all results as separate values, regardless of which extractor found them.
• Array: Groups repeated values into arrays, suitable for lists or collections.
• Key-Value: Pairs extracted keys (labels) with their corresponding values, ideal for forms or labeled data.
• OrderedArray: Combines results from multiple extractors in a specific order, such as address blocks.
• Pattern Based: Matches complex, multi-part patterns by coordinating multiple extractors.
• Other providers may support geometric, tabular, or region-based grouping.

Execution and Output

• The collation provider is applied after all extractors have run and produced their individual result sets.
• The provider merges, groups, or structures these results according to its logic, producing the final output for the Data Type.
• The choice of collation directly affects the structure of the extracted data (e.g., single values, arrays, key-value pairs, or complex objects).

Configuration

• Select the collation provider that matches your data extraction scenario from the drop-down menu.
• Configure any additional properties specific to the chosen provider (such as grouping rules, ordering, or pattern definitions).
• Collation is essential for scenarios where multiple extractors are used, or where the output must be structured in a particular way.

Advanced Usage

• Use collation to enforce relationships between extracted values, such as spatial proximity, order, or pattern conformance.
• For table extraction, use a collation provider that supports row and column grouping.
• For address or block extraction, use ordered or geometric collation to assemble multi-line or multi-field results.

Best Practices

• Choose the simplest collation method that meets your needs.
• Test extraction results with different collation providers to ensure correct grouping and output structure.
• Document the rationale for your collation choice, especially in complex scenarios.

Overview

The 'CultureFilter' property allows you to restrict extraction to documents with specific culture codes (locales). This is useful for handling documents in multiple languages or regional formats.

How It Works

• If the list is empty, extraction will run on all documents regardless of culture.
• If one or more cultures are specified, extraction will only occur for documents with a matching culture code.
• The culture code is typically determined by the document's metadata or by upstream processing steps.

Configuration

• Click the ellipsis button to select one or more cultures from the available list.
• Use this property to ensure that extractors are only applied to relevant documents, improving accuracy and performance in multi-lingual or international scenarios.

Advanced Usage

• Use culture filtering to avoid false positives or misinterpretation of data formats (such as dates, numbers, or currency) in documents from different regions.
• Combine with extractor logic that is tailored to specific languages, scripts, or regional conventions.
• In multi-lingual projects, configure separate Data Types for each supported culture and assign appropriate extractors and collation methods.

Best Practices

• Always specify culture filters when extraction logic is language- or region-specific.
• Review and update culture filters as your document set or supported languages evolve.

Can be one of the following types:

Overview

The 'Input Filter' property allows you to define a Value Extractor that limits where extraction occurs within the input document. Instead of running extraction logic against the entire document, extraction is performed only within the regions or segments matched by the input filter.

This is especially useful for:

• Focusing extraction on specific pages, sections, or regions.
• Improving performance by reducing the amount of content processed.
• Simplifying extraction logic by isolating relevant content.

How It Works

• When an input filter is specified, it is executed first against the input document.
• For each match produced by the input filter, the main extraction logic (local, child, and referenced extractors) is executed only within that matched region.
• Output instance indexes are automatically adjusted to reflect their position in the original document.

Configuration

• Select a Value Extractor (such as a pattern match or region extractor) to use as the input filter.
• Configure its properties to match the desired region, page, or section.
• If no input filter is specified, extraction runs against the entire input.

Advanced Usage

• Use input filters to restrict extraction to:
  • The first or last page of a document.
  • The top N lines of each page.
  • Specific labeled sections (e.g., "PERSONAL INFO").
  • Custom regions defined by regular expressions or zonal extractors.
• Input filters can be combined with other extractors for layered filtering and extraction.

Sample Regular Expressions

Here are some examples of regular expressions the might be used with Pattern Match to create input filters for common scenarios:

• Restrict to the first page of the document: ^[^\f]+
• Restrict to the last page of the document: [^\f]+$
• Restrict to the first 5 lines of the document: ^([^\r\n]+\r\n){5}
• Restrict to the top 3 lines of each page: (^|\f)([^\r\n]+\r\n){3}
• Restrict to the "PERSONAL INFO" section: \r\nPERSONAL INFO[^\f]+\r\nEDUCATION

Best Practices

• Use input filters to reduce noise and improve extraction accuracy.
• Test input filters thoroughly to ensure they match the intended regions.
• Document the purpose and logic of each input filter for maintainability.

Can be one of the following types:

Overview

The 'Exclusion Extractor' property allows you to define a Value Extractor that identifies regions or values to be excluded from the final extraction results. Any result that overlaps with a match from the exclusion extractor will be discarded.

This is especially useful for:

• Removing false positives that occur in known irrelevant regions (such as headers, footers, or watermarks).
• Excluding values that match certain patterns or appear in specific zones.

How It Works

• The exclusion extractor is executed after the main extraction logic.
• Each output instance is checked for overlap with any exclusion match.
• If an overlap is found, the result is removed from the output.

Configuration

• Select a Value Extractor (such as a pattern match, region, or list match) to use as the exclusion extractor.
• Configure its properties to match the regions or values you wish to exclude.
• If no exclusion extractor is specified, no exclusion filtering is performed.

Advanced Usage

• Use exclusion extractors to filter out repeated headers, footers, or other non-data content.
• Combine with input filters and subtraction extractors for layered filtering.
• Exclusion logic is especially helpful in documents with recurring noise or boilerplate text.

Best Practices

• Test exclusion extractors thoroughly to avoid removing valid results.
• Document the exclusion logic for future maintainers.

Can be one of the following types:

Overview

The 'Subtraction Extractor' property allows you to define a Value Extractor that identifies content to be removed from each output value after extraction. This is useful for cleaning up results by stripping unwanted substrings, such as labels, prefixes, suffixes, or formatting artifacts.

How It Works

• After extraction, the subtraction extractor is executed against each output value.
• Any content matching the subtraction extractor is removed from the value.
• If the resulting value is empty or contains only whitespace, the entire result is discarded.

Configuration

• Select a Value Extractor (typically a pattern match or region extractor) to use as the subtraction extractor.
• Configure its properties to match the content you wish to remove (e.g., trailing punctuation, labels, or formatting).
• The subtraction extractor must match a contiguous sequence of characters within the text flow.

Constraints

• The subtraction extractor cannot use Collation Provider methods that combine instances geometrically.
• It is intended for simple substring removal, not for complex region-based or multi-part extraction.

Advanced Usage

• Use subtraction extractors to remove labels (e.g., "Name: John Smith" → "John Smith").
• Clean up OCR artifacts or unwanted formatting from extracted values.
• Combine with exclusion extractors for comprehensive result filtering.

Best Practices

• Ensure the subtraction extractor is specific enough to avoid removing valid data.
• Document the purpose and logic of each subtraction extractor for maintainability.

Overview

The 'Lookup' property allows you to specify a Value Lookup object that validates each extracted value against a predefined list of known values (a vocabulary). This can be used to filter out invalid results, correct misspellings, or standardize output values.

How It Works

• After extraction (and after exclusion and subtraction filtering), each output value is checked against the lookup list.
• If a match is found, the value may be corrected or replaced according to the lookup configuration.
• If no match is found, the value can be discarded or flagged as invalid, depending on the lookup settings.

Configuration

• Click the ellipsis button to configure the vocabulary list and lookup options.
• Specify the list of valid values, correction rules, and any custom actions for unmatched values.
• Use lookup options to enable case-insensitive matching, fuzzy matching, or auto-correction.

Advanced Usage

• Use lookups to enforce data quality and consistency (e.g., standardizing company names, product codes, or categories).
• Correct common OCR errors or misspellings automatically.
• Combine with result filters and post-processing for robust data validation.

Best Practices

• Keep vocabulary lists up to date and relevant to your data set.
• Use lookup correction features judiciously to avoid introducing errors.
• Document the lookup logic and update procedures for future maintainers.

Can be one of the following values:

Deduplication ensures that only a single Data Instance is retained when multiple results overlap in the document content. This is especially useful when multiple extractors or extraction techniques may produce redundant or overlapping results.

Common Deduplication Scenarios

• Redundant Extraction:
  Multiple extractors may target the same value for redundancy. Deduplication ensures only one result is included in the output.
• Preference by Confidence:
  When some extractors are preferred, configure them to output higher confidence. Deduplication by confidence will favor these results.
• Self-Containing Values:
  When one result is a substring of another (e.g., "OWNERSHIP REPORT" vs. "MINERAL OWNERSHIP REPORT"), deduplication by length or area will retain the more specific match.

Usage

Set the deduplication mode using the 'Deduplication Mode' property. When enabled (not 'Disabled'), the 'Compare By' property is also exposed, allowing further control over how duplicates are detected.

Can be one of the following values:

The comparison mode controls how overlap or duplication is detected between Data Instances when deduplication is enabled.

The selected mode affects which results are considered overlapping and subject to deduplication, and is configured using the 'Compare By' property when deduplication is enabled.

The Result Filter allows you to configure a set of criteria that each Data Instance must meet to be included in the final result set. Results that do not match the specified conditions are excluded, enabling precise control over which values are retained for downstream processing or export.

Configuration and Usage

• Configure filter properties to target results by page, confidence, length, character composition, geometric region, font size, alignment, and more.
• Multiple filter criteria can be combined; a result must satisfy all enabled filters to be included.
• Use the 'Allowed Pages' property to restrict results to specific pages.
• Set 'Minimum Confidence' to exclude low-confidence results.
• Use 'Length Range' and 'Character Classes' to filter by value length and character makeup.
• The 'Exclusions' property allows you to specify a Value Extractor whose results will be used to exclude overlapping matches.
• Geometric filters (such as 'Region', 'Width Range', 'Height Range', and 'Font Size Range') allow you to target results by their position and size on the page.
• Emphasis filters (such as 'Character Casing', 'Case Similarity', 'Minimum Font Scale', and 'Require Underline') help isolate results with specific visual or typographic emphasis.

Typical Scenarios

• Restricting to a Region:
  Only include results that fall within a specific area of the page, such as a header or footer.
• Filtering by Confidence:
  Exclude results below a certain confidence threshold to improve data quality.
• Targeting Emphasized Text:
  Isolate headings, titles, or other elements with distinctive font size, casing, or underlining.
• Excluding Unwanted Results:
  Use the 'Exclusions' property to remove results that overlap with known false positives or irrelevant content.

Related Types

• Data Instance: The objects being filtered.
• Value Extractor: Used for exclusion filtering.
• Character Class Filter, Logical Rectangle, Logical Range, Horizontal Alignment Settings: Used for specific filter criteria.

For more information, see the documentation for Data Instance, Value Extractor, and Result Filter properties.

The Result Set Options class provides a flexible set of controls for shaping the output of data extraction and classification activities in Grooper. It allows you to define how individual results are adjusted, how the overall result set is filtered or ordered, and how output values are normalized for downstream use.

Overview

Use this class to:

• Normalize extracted values (e.g., unify variations, apply case or word transformations).
• Adjust or override confidence scores for results.
• Remove unwanted content from output values.
• Sort, filter, or deduplicate the result set.
• Enforce minimum/maximum hit counts or truncate the number of results.

These options are commonly configured on Data Fields, Data Types, or other extraction elements to ensure that the output meets business requirements and is ready for validation, export, or further processing.

Key Scenarios

• Value Normalization:
  Use 'Value Override', 'Case Normalization', or 'Word Normalization' to standardize output values, making them consistent for classification or export. For example, you can map multiple document title variations to a single normalized feature, or apply stemming to treat inflected word forms as equivalent.

• Confidence Adjustment:
  The 'Confidence Override' and 'Confidence Multiplier' properties allow you to boost or reduce the confidence of results, influencing their precedence in classification or extraction. You can also factor in character-level OCR confidence for more granular control.

• Result Set Shaping:
  Use 'Hit Count Range' to require a specific number of results, 'Distinct' to remove duplicates, 'Sort Order' to control result ordering, and 'Truncate At' to limit the number of returned results.

• Content Removal:
  The 'Subtractor' property lets you specify a Value Extractor to remove unwanted content (such as extraneous text or formatting) from each output value.

Processing Flow

When applied, the options in this class are processed in a defined sequence:

1. Individual Value Adjustments:
   Each result is normalized, cleaned, or transformed according to the configured options.
2. Result Set Operations:
   The set of results is filtered, sorted, deduplicated, or truncated as specified.
3. Type Enforcement:
   If a specific value type is required, results are converted and validated accordingly.

This ensures that the final output is both clean and conforms to the requirements of downstream consumers.

Usage Guidance

• Configure only the options relevant to your scenario; defaults are designed to minimize unnecessary processing.
• For best results, test your configuration with representative data to ensure that normalization, filtering, and confidence adjustments behave as expected.
• For more details on individual options, see the documentation for each property.

Can be one of the following types:

Overview

The 'PostProcessing' property allows you to define a Result Processor that applies additional logic to each output instance after extraction, filtering, and lookup operations have completed. Post-processing can be used to transform, normalize, or further validate extracted values before they are returned as final results.

How It Works

• After all extractors, filters, and lookups have been applied, the post-processing logic is executed for each output instance.
• The Result Processor can modify the value, confidence, geometry, or other properties of each result.
• Post-processing can also be used to enforce formatting rules, perform calculations, or apply custom business logic.

Overview

The 'Group Names' property provides a comma-separated list of all named groups produced by the configured extractors and collation provider. Named groups are typically defined in regular expressions or extraction patterns and represent subfields or components of the extracted data.

How It Works

• This property is read-only and automatically aggregates group names from all extractors and the collation provider.
• Group names are used to map extracted values to specific fields, columns, or data elements in the output.
• The display is intended for informational purposes, helping users understand which subfields are available for mapping or reference.

Usage

• Use this property to verify that all expected named groups are present in the extraction configuration.
• Reference group names when mapping extracted data to Data Fields, Data Columns, or other data elements.
• Useful for debugging and documentation of complex extraction logic.

The 'ExtractorCount' property summarizes the number and types of extractors configured for this Data Type. It includes:

• The local extractor (if set)
• All direct child extractors (such as Value Readers, Field Classes, or child Data Types)
• All referenced extractors (from the 'Referenced Extractors' property)