[{"keyword": "@","description": "A prefix that need to be used with a required ontology field or a table name for referencing them in an ontology rule.","synonyms": []},{"keyword": "anchor","description": "A value available in a document that provides the spatial reference of an ontology field in respect to itself (like 'below', 'left', and other).","synonyms": []},{"keyword": "audit trail","description": "A trail of activities done by an authorized user that have been logged for audit purpose.","synonyms": []},{"keyword": "baseline","description": "An activity that measures the accuracy of the extracted data against the golden data. Golden data refers to the data that is considered to be fairly accurate.","synonyms": ["baselining"]},{"keyword": "batch of documents","description": "Often used in context of documents; it represents a set of documents used for the purpose of baselining and training OCR- and ML-based models.","synonyms": []},{"keyword": "blacklist profile","description": "A set of keywords that prevent a document from being taken up for data extraction.","synonyms": []},{"keyword": "class library","description": "A DLL that can be added to the application to include methods for data extraction.","synonyms": []},{"keyword": "cleanup","description": "Refers to the activity of cleaning-up extracted data (such as removing leading and trailing spaces, etc.) using the available clean-up rules.","synonyms": ["clean-up"]},{"keyword": "confidence","description": "The level of certainty that the OCR system has regarding the accuracy of the recognized characters. The confidence level range is 0-9, where 0 represents least confidence and 9 represents highest confidence.","synonyms": []},{"keyword": "copilot","description": "An ML-based capability for automatically generating ontology.","synonyms": []},{"keyword": "customer keys","description": "A set of public and private keys that a customer can use to encrypt a request sent to a TruCap+ API and decrypt the subsequent response sent by the API.","synonyms": []},{"keyword": "datasource","description": "Represents the data stored in a database (either as a static list or as a table) that can be looked-up for populating ontology fields.","synonyms": ["data source"]},{"keyword": "datatable","description": "A datatable comprises of a table that is generated for a 'Datatable' type ontology field and is populated by looking-up data from the database.","synonyms": ["Data Table"]},{"keyword": "document category","description": "Depending on the context, a category may either imply 1) the first level of document categorization or, 2) refer to the labels applied to business keywords for the purpose of source identification.","synonyms": []},{"keyword": "document hierarchy","description": "A document hierarchy in TruCap+ is represented by 'Category', followed by 'Project', and then  by 'Subtype'.","synonyms": []},{"keyword": "document ontology","description": "It is a means through which important fields in a document can be identified for the purpose of data extraction. It also defines the methodologies that are to be used for extracting a given field.","synonyms": ["ontology"]},{"keyword": "document subtype","description": "The last level of document categorization after 'Category' and  'Project'.  It is usually used to represent the nature of a document set (like invoices, reports, insurance forms, and others).","synonyms": ["subtype"]},{"keyword": "document template","description": "A template is a format in which data is arranged in a document. Based on a template, an ontology is defined for the purpose of extracting data from the documents.","synonyms": []},{"keyword": "enqueue","description": "A feature that allows high-priority documents to be processed on-demand by bypassing the queued documents.","synonyms": []},{"keyword": "field type","description": "Also known as datatype, it represents the nature of an ontology field. Ontology fields are broadly classified as standard type (like numeric, alphanumeric, and others) and business type (like email, website, and others).","synonyms": []},{"keyword": "footer section","description": "For extraction purpose, the fields in an ontology are categorized under header, details, and footer sections. The 'details' section generally refers to the document table, and  the 'header' and 'footer' sections refer to the document data above and below the table respectively.","synonyms": []},{"keyword": "golden batch","description": "It refers to the batch of data that has been manually inspected and quality-checked for accuracy and is used as a benchmark to assess the accuracy of other OCR- and ML-based models.","synonyms": ["golden","golden data"]},{"keyword": "golden data","description": "It refers to the batch of data that has been manually inspected and quality-checked for accuracy and is used as a benchmark to assess the accuracy of other OCR- and ML-based models.","synonyms": ["golden","golden batch"]},{"keyword": "header section","description": "For extraction purpose, the fields in an ontology are categorized under header, details, and footer sections. The 'details' section generally refers to the document table, and  the 'header' and 'footer' sections refer to the document data above and below the table respectively.","synonyms": []},{"keyword": "heterogeneous documents","description": "Represents a set of documents belonging to different subtypes like invoice, report, insurance form, etc. ","synonyms": []},{"keyword": "HITL","description": "Human-in-the-loop (HITL) is a term used to represent human interaction with the data extraction process to provide data for the unextracted fields and doing quality-check on the data.","synonyms": []},{"keyword": "image profile","description": "A set of attribute values that can be applied to an image for improving it's quality (contrast, brightness, etc.)  for the purpose of data extraction.","synonyms": ["image profiles"]},{"keyword": "input location","description": "The source location where the documents from which the data is to be extracted are kept. TruCap+ offers wide variety of input location choices to its users like FTP, cloud, shared location, etc.","synonyms": []},{"keyword": "input source","description": "The source location where the documents from which the data is to be extracted are kept. TruCap+ offers wide variety of input location choices to its users like FTP, cloud, shared location, etc.","synonyms": ["input location"]},{"keyword": "integrated document","description": "Documents belonging to different subtypes combined as one for business purpose. ","synonyms": []},{"keyword": "integrated documents","description": "Documents belonging to different subtypes clubbed together in an integrated project for business purpose. (Such documents are heterogeneous by nature.)","synonyms": ["integrated document","integrated project","heterogeneous"]},{"keyword": "integrated project","description": "A type of project for managing heterogenous documents.","synonyms": []},{"keyword": "keyword annotation","description": "An activity of providing additional information in regard to logos and business keywords for the purpose of source identification that leads to populating of ontology fields for a give document.","synonyms": ["annotation"]},{"keyword": "Knowledge master","description": "Knowledge master represents a collection of database tables that represent information about document source (owner) such as name, address, office, etc. This data is used to populate ontology fields after the document source is identified.","synonyms": []},{"keyword": "leadtools","description": "Refers to the LEADTOOLS software development toolkits created by LEAD Technologies, Inc. to integrate functionalities such as OCR (Optical Character Recognition), barcode recognition, etc.","synonyms": ["lead tools"]},{"keyword": "logo identification","description": "Logos are used for source identification that leads to data being fetched from the database rather than being extracted using OCR, which is comparatively slower and less accurate.","synonyms": []},{"keyword": "data lookup","description": "An approach in which data is fetched directly from the database rather than using OCR-based extraction, which is comparatively slower and less accurate.","synonyms": []},{"keyword": "manual handling","description": "An exercise carried out by authorized operators to manually inspect the extracted data and add or modify it as per their discretion.","synonyms": ["MI"]},{"keyword": "master data","description": "A set of data maintained in a database for purpose of populating ontology fields and for verifying the correctness of extracted data. This data is fed manually by an authorized user.","synonyms": []},{"keyword": "ontology field","description": "An ontology field represents the data in a given document that needs to be extracted by using the methodologies defined for it in the ontology.","synonyms": ["ontology fields"]},{"keyword": "output file","description": "A file that contains the extracted data.","synonyms": []},{"keyword": "output source","description": "A location where the output files containing the extracted data are stored.","synonyms": []},{"keyword": "primary customer","description": "Primary customer are official partners of Datamatics that can sell and manage TruCap+ subscriptions.","synonyms": []},{"keyword": "project template","description": "In the context of customized DLLs, a project template defines the interface for an extraction method, which a user can use to define a customized extraction rule.","synonyms": []},{"keyword": "published model","description": "A model that has been trained and approved for the purpose of data extraction and related activities.","synonyms": []},{"keyword": "quality check","description": "An activity carried out by authorized users to check the accuracy of data that has been either programmatically extracted or filled-in by a manual operator.","synonyms": ["QC"]},{"keyword": "queues","description": "In context of TruCap+, a queue represents one of the several stages of processing in which documents are queued awaiting their turn to get processed.","synonyms": ["queues"]},{"keyword": "regex","description": "A regex type is a regular expression that is created to identify and extract business data like phone numbers, websites, zip codes, etc.","synonyms": ["regular expression"]},{"keyword": "RoI","description": "Region of Interest (RoI) is an area that is marked by a user on a document for identifying the region where a data field is most likely to be found.","synonyms": ["region of interest"]},{"keyword": "ontology rule","description": "Rules are created to extract data and process it as per the given requirement. Different categories of rules are available like extraction rules, clean-up rules, and so on.","synonyms": []},{"keyword": "source identifier","description": "Source identification leads to the population of ontology fields by fetching data directly from the database which is relatively faster and more accurate than OCR based extraction.","synonyms": ["source identification"]},{"keyword": "synonym","description": "Keywords and synonyms are often used interchangeably to represent a group of labels that often appear in documents in vicinity to a relevant data field. Using a keyword and a spatial reference to it (like above or below the keyword), it becomes possible to extract the data.  For example, 'Number', 'No.', and 'Sno.' are possible keywords for serial number. ","synonyms": ["synonyms"]},{"keyword": "table field","description": "An ontology table represents the tabular data in a given document that needs to be extracted by using the methodologies defined for it in the ontology.","synonyms": ["ontology table","ontology tables"]},{"keyword": "training workbench","description": "It is a centralized dashboard which displays the ML models published and available for training","synonyms": []}]