Glossary

Asset Description Metadata Schema is a standardized metadata vocabulary, defined by the W3C to describe semantic assets on the web defined as highly reusable metadata (e.g. xml schemata, generic data models) and reference data (e.g. code lists, taxonomies, vocabularies). All pertaining data in a PoolParty project will be stored in a separate ADMS graph.

An alternative lexical label for a concept which may include synonyms, near-synonyms, abbreviations, acronyms, spelling variations or variant phrases with the same meaning. Alternative labels support tagging, search, and retrieval by different names for the same concept. In SKOS, alternative label is a standard property of a concept expressed as skos:altLabel.

Stands for Application Programming Interface, a software component which, as a set of rules and protocols, serves as an intermediary connecting two different software applications so that they can share data or functionality. As middleware, PoolParty has various APIs to connect to both front and backend applications, supporting tagging and searching. Examples of PoolParty APIs include for instance Extractor APIs, Thesaurus Ontology Management API, Entity Extractor APIs, UnifiedViews APIs, Recommender APIs etc.

In PoolParty Thesaurus Management is a feature to manage approval of additions or changes to a thesaurus project by multiple users. It is an approach to collaborative thesaurus management, status control and, release cycle management.

A relationship between two concepts in a thesaurus or taxonomy, whereby the concepts are related to each other in some way that is not hierarchical, rather a bidirectional symmetrical relationship. Another use for this property is between two categories where neither is more general or specific than the other. The "Related Concept" relationship is used to express this, which is designated in SKOS as skos:related. However bear in mind that 'related concept' is not the only way to express an associative relation as there can also be more specific custom relations.

A characteristic of on an entity, such as a concept or member of a class. In an ontology, an attribute is also called a datatype property. Attributes are like metadata on concepts. Notes and definitions of a concept could be considered attributes of the concept, but usually "attribute" refers to a "custom attribute," which is created in ontology management. Custom attributes support the inclusion of additional data in a PoolParty thesaurus project. Attributes are more common on named entities such as biographical details on persons names or corporate data on company names. Attributes can be of various types (text, numbers, Boolean expressions etc.)

In PoolParty, the concept label lookup feature which provides suggested matches after just three or more characters are entered into the search box. Autocomplete is available for general searches for concepts, searching for concepts when adding relations, and in Linked Data Administration linking to concepts in DBPedias well as using GraphSearch.

In PoolParty, the function to automatically add a set of concepts to a thesaurus is based on connecting to a Linked Open Data source, either from those available by default or those that you designate. The autopopulate function enables generating a seed thesaurus from harvested linked data, categorizing concepts, and autopoplulation based on categories.

A contextual designation for a concept in a hierarchical relationship that directs from one concept to another, where the latter has a broader meaning, is more generic, or is a larger whole with respect to its narrower parts. In SKOS, this is designated with skos:broader for use within a concept scheme and skos:broadMatch for use across different concept schemes or taxonomy projects.

A proposed concept that needs to be approved before it can be used in tagging. It is however not required to create concepts initially as candidate concepts. Candidate concepts are created either because a workflow has been set up that initially designates all concepts as candidates or because the concepts came from corpus analysis suggestions.

The task of assigning a grouping category for something. Categorization of documents is a capability of the PoolParty Extractor handled by the categorization service API call. It identifies categories which, defined as the top concepts in a thesaurus, are relevant to a document. This is done by identifying the concepts in the document and then finding the concepts’ top concepts in the thesaurus. Additionally, categorization of concepts is a capability of the Autopopulate function in PoolParty Thesaurus Management after linking to a DBPedia Categories source, which will suggest categories.

An administrative note that may be added to a concept to explain a change that has been made to other taxonomy editors, such as a change in a broader concept relationship or a definition. An explanation of why the change was made may be included, whereas the History log does not include explanatory notes. Change notes are one of three Note types that can be added to a concept from the Notes tab in PoolParty. It is one of the SKOS documentation elements and is expressed as skos:changeNote. It is similar to a History note, but intended for internal use only.

In an ontology, a named set of entities (individuals or subclasses) that share the same properties. Classes may exist in a hierarchy with subclasses. In a custom ontology, classes may correspond to selected concepts, top concepts, or concept schemes in linked SKOS taxonomy projects. Classes are created in ontology management, and the first step in creating an ontology is to create classes.

Assigning a category or a classification to a document or digital content item so that content can be consistently classified for analysis or findability. Typically, each item can receive only a single classification. Classification is to describe the document as a whole, rather than identifying multiple prominent concepts, as is done with tagging. In PoolParty, classification is automated with the Semantic Classifier.

One of the four types of SKOS mapping relations available when linking individual concepts across two linked taxonomy projects. Close match (skos:closeMatch) may be used to indicate that two concepts are sufficiently the same in meaning for the use case for linking the two taxonomies, but the meanings are not exactly equivalent and may not be appropriately equivalent in other use cases or contexts.

In SKOS, an additional optional way to designate a grouping of concepts for a specialized purpose, independent of any hierarchy, such as to be used in only specific implementations for example. In PoolParty one can create collections, give them names, and assign concepts to collections.

The basic unit in a SKOS taxonomy, which represents a thing, idea, or shared understanding of something. It is the combination of both a preferred label and its various alternative labels. A concept is more than just a term. Concepts can be identified with URIs, labeled with lexical strings in one or more natural languages, documented with various types of note, semantically related to each other, and aggregated into concept schemes.

An organized collection of concepts in a SKOS vocabulary. A concept scheme is a single controlled vocabulary, thesaurus, hierarchical taxonomy, facet within a faceted taxonomy or metadata property within a larger metadata scheme. Broader, narrower, and related relations may exist between concepts within a concept scheme.

Assigning of concepts to a content item (document, document section, page, digital asset, file, etc.), from among the concepts of a taxonomy/thesaurus, with focus on the concepts and their meaning and not terms or strings of text.

A body of multiple text documents, which may be used in the corpus analysis feature of PoolParty.

Stands for Comma Separated Value, which is a generic (non-proprietary) format for tabular data. "Tabular" means in a table of rows and columns. Spreadsheet software can export to the CSV format. PoolParty can import and export taxonomy and custom scheme data in Excel and CSV.

An ontology that someone creates newly for a specific purpose, rather than reusing an existing ontology. Most ontologies are custom ontologies, which could be completely original or could reuse classes and properties from existing ontologies. In PoolParty, all taxonomies that the PoolParty user creates are custom ontologies, in contrast to core taxonomies that are pre-loaded, which are more generic and are called Core ontologies. Custom ontologies also tend to be domain ontologies. Imported ontologies could be custom ontologies or domain ontologies.

In an ontology, an object type property of a class and its members, which is a semantic relation with another class and its members. The ontologist defines the relations, which are often specific for a class, although they could also be reused among different classes. The relations may have inverses/reciprocals (the relation name in the other direction), be directional in one direction only, or may on rare occasions be symmetrical (same in both directions).

A knowledge model in PoolParty based on one or more ontologies or parts of ontologies which is applied to one or more taxonomy projects to semantically enrich the taxonomy so that it has the additional features of an ontology, especially semantic relations and custom attributes. Custom schemes are managed in Ontology Management and are created based on existing ontologies. They may match a single ontology exactly.

Simply put, data harvesting is the process of extracting valuable data from a given source. This data can then be used for a variety of purposes, such as marketing or research. Data harvesting is extracting valuable data from target sources and putting them into your database in a structured format.

Data mining is the analysis of large sets of data in order to derive trends, whereas data harvesting is the process of extracting data from online sources to then build analyses. While data mining focuses more on the analysis of data, the focus of data harvesting is on the collection of data itself.

A PoolParty component validating whether (imported) data is compliant with RDF and SKOS standards as well as PoolParty standards.

DBpedia is a project aiming to extract structured content from the information created in the Wikipedia project. This structured information is made available on the World Wide Web.

A glossary or dictionary-type definition of the meaning of a concept. In PoolParty, the Definition field is one of the basic SKOS properties, along with Scope note which displays under the labels. It is classified as one of the documentation elements in SKOS and it is expressed as skos:definition.

An optional constraint that can be made on a class in ontology, whereby all members or instances of the class cannot belong to any other class. Reasoning can then be made based on class disjointedness.

Domain refers to the subject of a triple in an ontology. When creating custom relations and attributes in an ontology you may restrict the domain to specific classes, any concepts, any concept schemes or labels.

Stands for Data Processing Unit. In PoolParty, DPUs are modules used in UnifiedViews where they are linked to create pipelines allowing for example the extraction or loading of data. It is a persisted entity comprised of its properties, .jar file, and default configuration. Generally, a DPU takes input data, processes it, and saves the output to a specific data structure. The DPU types are Extractor, Transformer, Loader, and Quality.

An administrative note that may be added to a concept to explain to other taxonomy editors something other than a change, such as administrative work in progress or needing to be done on the concept. Editorial notes are one of three Note types that can be added to a concept from the Notes tab in PoolParty. It is one of the SKOS documentation elements and is expressed as skos:editorialNote.

Elasticsearch is a search engine based on the Lucene library. It provides a scalable, distributed and multitenant-capable search engine with a HTTP web interface and schema-free JSON documents. It is one of the world’s leading free and open search and analytics solutions. GraphSearch works with Elasticsearch among other search and storage backends.

One of the four types of SKOS mapping relations available when linking individual concepts across two linked taxonomy projects. Exact match (skos:exactMatch) is used to link two concepts which have identical preferred labels, minor variations in plural or capitalization, or are exact meaning synonyms in all cases.

A form of additional information about a concept by providing example usage or example sub-types. It is similar to Scope note. In PoolParty, the Example field is one of the extended SKOS properties, which is displayed when the user’s SKOS View is set to Advanced. It is classified as one of the documentation elements in SKOS, and it is expressed as skos:example.

In PoolParty Thesaurus Management, the designation of a concept for exclusion from tagging with the Extractor. In Corpus Management, the designation of an extracted term for exclusion from candidate concept status in future executions. Previously referred to as 'blacklist'.

In PoolParty an API-based model for extraction and thus the base for all functionality of the PoolParty Extractor. This model needs to be refreshed when for example new concepts have been added.

The PoolParty Extractor provides a highly performant, secure and clusterable semantic extraction service (in the cloud or on-premise) able to automatically extract the most relevant concepts, terms, and named entities from a given document or text section. The PoolParty Extractor automatically and accurately analyses documents and texts and then extracts meaningful phrases, named entities, categories or other metadata. Different data or metadata schemes can be mapped to a SKOS thesaurus used as a unified semantic knowledge model.

A database using graph structures for semantic queries with nodes, edges, and properties to represent and store data. A key concept of the system is the graph. A graph database stores nodes and relationships instead of tables or documents. Data is stored just as you might sketch ideas on a whiteboard.

In PoolParty GraphEditor is an additional component to edit knowledge graphs.

In PoolParty GraphSearch API incorporates GraphSearch into your existing search applications. PoolParty GraphSearch can use Solr, Elasticsearch or a graph database as a search index. GraphSearch functionality encompasses amongst others semantic search as well as concept and facet based document filtering.

An add-on module to PoolParty with a separate installation and configuration procedure. It provides a graphical view of a taxonomy + custom scheme project allowing the user to interact and expand different branches.

In SKOS, a kind of variant label of a concept which is known instead by a preferred label, but is distinguished from standard alternative labels by being designated as not to display to end users. It may be used for misspellings, deprecated labels, slang or offensive language labels, or labels intended to match to text for auto-tagging but not intended to match user entered search strings.

A group of concepts interlinked by hierarchical relationships. Most controlled vocabularies of subjects contain multiple hierarchies that may be of various sizes.

A note that may be added to a concept to explain a significant change that has been made, such as change in a preferred label with slight change in meaning. History notes are one of three Note types that can be added to a concept from the Notes tab in PoolParty. It is one of the SKOS documentation elements and is expressed as skos:historyNote. It is similar to Change note, but could be designated separately for display to users of downstream applications, not just other PoolParty taxonomy editors.

In SKOS, the relation between a concept and what concept scheme it belongs to. It is expressed as skos:inScheme. When using the PoolParty Thesaurus Management tree user interface a concept is by default in the concept scheme of its hierarchy, and the in scheme relation is not indicated. However, there may be times when you want to designate a different concept scheme or have a pipeline make use of In scheme designations, and then you may use the inScheme Settings. By using the inScheme property it is possible for the same concept to be linked to several concept schemes.

Java Enterprise Edition platform

A type of graph-based knowledge base that typically comprises both a knowledge organization system of an ontology plus a taxonomy or other controlled vocabularies and also specific instance data that is interconnected.

A name or words that describe a concept. Labels may be preferred, alternative, or hidden. Labels may or may not be displayed to users but in either case their text strings may be matched to user searches or to text strings or content. Labels for the same concept may be in different natural languages.

A large language model (LLM) is a computational model notable for its ability to achieve general-purpose language generation and other natural language processing tasks such as classification. Based on language models, LLMs acquire these abilities by learning statistical relationships from vast amounts of text during a computationally intensive self-supervised and semi-supervised training process. LLMs can be used for text generation, a form of generative AI, by taking an input text and repeatedly predicting the next token or word.

Data published on the web in accordance with W3C specifications for web publishing that makes the data part of the Semantic Web. It extends the World Wide Web by making shared data machine-readable. Linked data follows HTTP protocols, has URIs (uniform resource identifiers), and is structured according to RDF.

A web-browser view of a taxonomy/thesaurus managed in PoolParty, which is however external to the PoolParty application and has its own URL (in the form servername.poolparty.biz/projectname.html). It is externally accessible by non-PoolParty users. Users of the Linked Data Frontend may browse or search the taxonomy, follow relationship links to other concepts, and query the taxonomy via a SPARQL endpoint. When a PoolParty project is designated as “public” (assigned to the Public user group) it is available in the Linked Data Frontend.

A PoolParty feature to create a starter or “seed” thesaurus based on extracting (or “harvesting”) concepts, their category hierarchy, and optionally alternative labels, definitions and other relations from linked open data source of DBPedia. The DBPedia Categories lookup can be activated in the Linked Data Sources of the Advanced menu in Thesaurus management.

Linked data (data published on the web in accordance with W3C specifications) that has no restrictions on access, so it is available free for reuse. Also referred to as LOD.

The use and development of computer systems able to learn and adapt without following explicit instructions by using algorithms and statistical models to analyze and draw inferences from patterns in data. Machine learning-based named-entity recognition capabilities are integrated within the PoolParty Extractor for auto-tagging. PoolParty’s Semantic Classifier includes various machine learning algorithms to select from for document auto-classification.

Standardized data of the importance of designated types for a collection of resources, assets, documents, files, or graphs. It is sometimes referred to as 'data about data'. Metadata refers to the combination of type/property/element (such as title, creator, date, language, location, document type, audience, and subject) and the specific values for each property. Values for some metadata properties, such as subject or document type, are best managed as controlled vocabulary concepts. Controlled vocabulary metadata can be managed in PoolParty Thesaurus Management with a concept scheme for each metadata property and concepts for the controlled values, whereas metadata with uncontrolled values (such as text strings) can be be managed in PoolParty Ontology Management as attributes. A PoolParty project itself is a resource with metadata which, when the project name is elected, can be viewed and edited in Thesaurus Management under the tab for Metadata & Statistics.

Metadata of all projects on a PoolParty Thesaurus server which is stored in a system repository as a graph database for the PoolParty Thesaurus server. This is not the same as the actual product data for a project which is stored in a separate project repository.

A subset of data in a triple store database that has been given a unique label (name). A graph database can contain any number of named graphs alongside its default graph. Having multiple RDF graphs in a single document/repository and naming them with URIs provides useful additional functionality. PoolParty project repositories contain several named graphs.

A contextual designation for a concept in a hierarchical relationship that directs from one concept to another where the latter has a narrower meaning, is more specific, or is a part with respect to a systematic whole. Its inverse is broader concept. In SKOS, this is designated with skos:broader for use within a concept scheme and skos:broadMatch for use across different concept schemes or taxonomy projects.

Technologies to automatically discern the meaning of text based on both linguistics and computer science, typically involving the parsing of text so that the grammar is also taken into consideration. NLP technologies are often used in entity extraction and auto-tagging. The PoolParty Extractor uses NLP for auto-tagging.

An optional additional label-like field on a concept that is intended for use to supplement a preferred label. Generally, it is used for alphanumeric codes such as classification numbers in classification systems. In PoolParty, the Notation field is one of the extended SKOS properties displayed when the user’s SKOS View is set to Advanced. It is expressed as skos:notation.

A knowledge organization system aiming at describing a domain of knowledge. An ontology is semantically richer than a taxonomy or thesaurus by specifying classes of things (corresponding to SKOS concepts), semantic relationships between classes and their members, and the attribute types available to classes and their specific attributes for individual members (concepts). Ontologies are built on W3C standards of RDF-Schema and OWL (Web Ontology Language). In PoolParty, ontologies are applied to taxonomy projects via an intermediate knowledge model called a custom scheme.

The series of steps that allows data from one system to move to and become useful in another system. A data pipeline works by pulling data from the source, applying rules for transformation and processing, then pushing data to its destination in a process known as Extract-Transform-Load (ETL). UnifiedViews is the PoolParty module for managing the pipeline and ETL process. The pipeline is represented as a graph with data processing units (DPUs) as nodes and connecting lines (edges) as data flow. A common pipeline involves the PoolParty Extractor for tagging documents.

The file saved and downloaded when you select to download a project, in contrast to exporting a tabular or RDF project. It has the file extension *.ppar (which stands for PoolParty archive), and can be uploaded again to the same or different server to copy a project or create project that is based on an existing project. In addition to all thesaurus data (concepts, relationships, notes, etc.) the archive may include candidate concepts, the history log, ontologies and custom schemes, and corpora used in corpus analysis. This is all downloaded as a single compressed archive file, which can then be uploaded again to the same or different PoolParty server.

A designated label for a concept by which the concept is known and that is shown in hierarchical and other displays (such as a type-ahead search) of a taxonomy. In SKOS, preferred label is a standard and mandatory property of a concept and it is expressed as skos:prefLabel.

In PoolParty this is the designation for a knowledge organization system or knowledge model which could be a taxonomy, depending on how that is defined, in combination with one or more applied custom schemes. A single project is usually created for a single use case involving tagging a shared set of content for a defined set of users. A project may comprise multiple concept schemes which may be different vocabulary types (term list, name authority, hierarchical taxonomy, thesaurus) as well it can also include one or more corpora. PoolParty user rights can also be set at the project level.

Retrieval augmented generation (RAG) is a natural language processing (NLP) technique that combines the strengths of both retrieval- and generative-based artificial intelligence (AI) models.

The object of a triple in an ontology. When creating custom relations and attributes in an ontology, you may restrict the range to specific classes, concepts, concept schemes, or labels.

Stands for Resource Description Framework which is a standard supported by the W3C for representing information about resources on the World Wide Web to make it machine-readable and interchangeable. In RDF, everything is expressed as subject-predicate-object triples, and every resource in a triple has a URI.

In a front-end application the suggestion of relevant related content which is not directly what the user searched for. PoolParty enables the building of knowledge-based recommendation applications comprised of a taxonomy and ontology/custom scheme in the Thesaurus Management, GraphSearch and the recommender plug-in.

A generic associative relationship between two concepts in a thesaurus which is not hierarchical but deemed to be sufficiently related. It is symmetrical, the same in both directions. It is recommended for use within a single concept scheme. As a SKOS relation type it is expressed as skos:related.

One of the four types of SKOS mapping relations available when linking individual concepts across two linked taxonomy projects. Related match (skos:relatedMatch) may be used to indicate that two concepts would have an associative relationship between them, like skos:related, but they belong to different taxonomy/Thesaurus projects or different concept schemes.

Data in a relational database where data is structured in tables of rows and columns. Since PoolParty manages data in a graph database, data sources that are relational need to be converted, which can be done with UnifiedViews.

Release notes are documents distributed with software products or hardware products, sometimes when the product is still in the development or test state. For products that have already been in use by clients, the release note is delivered to the customer when an update is released.

An optional note on a concept in a thesaurus to clarify usage, such as by restricting the scope of the concept. It is similar to but not the same as a definition of a concept. In PoolParty, the Scope note field is one of the basic SKOS properties, along with definition, which displays under the notation. It is classified as one of the documentation elements in SKOS, and it is expressed as skos:scopeNote.

Scrum is an agile project management framework, commonly used in software development, although it has been used in other fields including research, sales, marketing and advanced technologies.

The area of artificial intelligence that deals with comprehending the meaning of language. Semantic AI, which includes both natural language processing (NLP) and natural language understanding (NLU), provides a framework to perform tasks automatically, utilizing various machine learning and logic-based approaches and also utilizing background knowledge often stored in knowledge graphs. Thus, the knowledge graph can also explain the AI, and as such Semantic AI is considered an application of Explainable AI. PoolParty Extractor and Semantic Classifier utilize Semantic AI.

An optional add-on auto-classification module to PoolParty Enterprise Server, Search Server, or Semantic Integrator. It classifies documents by combining semantic knowledge graphs with machine learning. It offers a choice of algorithms for classification, such as SVM, Deep Learning, and Naive Bayes.

A relationship between pairs of concepts that has “meaning.” SKOS defines its thesaural relationship types (broader, narrower, and related) as “semantic relations,” but usually semantic relationships refer to those that are custom created and defined by the taxonomist or ontologist and have more meaning than the generic thesaural relationships of broader, narrower, or related. These are the relations created in Ontology Management.

An extension of the World Wide Web to apply meaning to data and links to make data on the web machine-readable and interoperable. This supports linked meaningful data, which on the open web is Linked Open Data, instead of merely linking pages without context. It has been a project of the W3C which has published various guidelines and standards to support the Semantic Web, including RDF, RDF-Schema, SKOS, OWL, and SPARQL. Semantic Web standards may also be adopted within an enterprise to enable interoperability of data, independent of a specific software vendor. PoolParty manages data according to Semantic Web standards.

A concept in a PoolParty thesaurus that is identified in corpus analysis as relevant from a document contained in a corpus, even when it is not explicitly included in the text. Shadow concepts are identified on the basis of co-occurrence with other concepts.

Stands for Simple Knowledge Organization System, a W3C recommendation for a standard data model for controlled vocabularies including thesauri, classification schemes, taxonomies, and subject-heading systems. As a part of the Semantic Web family of standards based on RDF, SKOS has become the standard data model for sharing and linking knowledge organization systems. Vocabularies created in PoolParty Thesaurus Management are fully SKOS-based.

Stands for SKOS eXtension for Labels, a W3C recommendation that is an addendum to SKOS by specifying when and how labels can be treated as resources, and thus have properties of their own. It does this by adding three more elements not found in SKOS: label, label relation, and literal form. SKOS-XL is supported in PoolParty as an add-on/upgrade feature.

In PoolParty, this is a database backup feature available for Admin and Superadmin user roles. A backup can be restored from a snapshot, either made by the system automatically or periodically, or made manually before changes.

An open-source enterprise-search platform developed by the Apache Software Foundation, written in Java. Its major features include full-text search, hit highlighting, faceted search, real-time indexing, dynamic clustering, database integration, NoSQL features and rich document handling. PoolParty’s own search module, PoolParty Search Server (PPS), is based on Solr.

Stands for SPARQL Protocol and RDF Query Language, which is a standard query language and data access protocol for use with data in the Resource Description Framework (RDF), specific by the W3C.

One of the two main approaches to AI, which deals with making predictions and finding patterns in structures of data to make those predictions, based on statistical data. Statistical AI is used in machine learning, and in PoolParty it is used for document classification.

The full hierarchy of narrower concepts and their narrower concepts, etc. from the point of a designated concept or top concept. Certain functions in PoolParty Thesaurus Management can be applied to a subtree, which means involving the full hierarchy narrower concepts to the selected concept. These include importing, exporting, deleting, applying or removing a custom class, and applying or removing in scheme relations. You can also move complete subtrees to a new position within the hierarchy, for instance to a different top concept.

One of the two main approaches to AI, which deals with logic and reasoning. Symbolic AI uses reasoning by “symbols” which are generally the variables in logical statements. Knowledge graphs can play a role in symbolic AI.

The process of adding semantic metadata (especially from a controlled vocabulary, taxonomy, or thesaurus) to enrich content for purposes of improved findability, comparison and analysis and content lifecycle management. Tagging can be done manually or with automated methods based on entity extraction and natural language processing. The PoolParty Extractor enables auto-tagging.

A group of controlled vocabulary terms or concepts organized into a hierarchical structure. A set of term lists organized by aspect/facet used in combination for filtering search results may also be considered a taxonomy, as a 'faceted taxonomy'. There is no strict definition of what constitutes a single taxonomy; it could be a SKOS concept scheme, or it could be multiple concepts schemes managed together as a project in PoolParty.

A word or phrase that has meaning. Non-SKOS taxonomies or thesauri refer to terms, rather than concepts, but 'term' is somewhat ambiguous, because it can mean either label or concept so it does not appear in the SKOS standard, which instead refers to concepts and labels. In PoolParty, terms are what is extracted from text with the PoolParty Extractor, and they may then be matched to concepts for tagging, or in corpus analysis the terms may be added as candidate concepts to a thesaurus.

A structured type of controlled vocabulary which provides information about each concept and its relationships to other concepts within the same thesaurus. Hierarchical (broader/narrower) and associative (related) relationships and scope notes are common features. Unlike taxonomies, thesauri do not have a limited number of top concepts. Published standards (ANSI/NISO Z39.19 and ISO 25964-1) provide best practices for creating thesauri.

In a thesaurus or taxonomy, a concept that has no broader concepts, only narrower concepts. It is the top/highest concept of its own hierarchy. In SKOS, top concept is managed as predicate (like a relation), not a type of concept, with the designations skos:hasTopConcept and the inverse skos:topConceptOf between a concept and its concept scheme. In PoolParty, top concepts are designated in the taxonomy tree in dark green.

A statement comprised of three parts: subject-predicate-object, and the foundation of the Resource Description Framework (RDF) data exchange model, and in turn the data model for taxonomies, thesauri, ontologies and other knowledge organization systems based on RDF and its derived standards from the W3C. All data in PoolParty is stored as triples which can be viewed. Triples can be queried and edited with the SPARQL query language.

A graph database that stores data in RDF (Resource Description Framework) triple statements. Graph databases tend to be of either two types: triple stores or labeled property graphs. PoolParty stores its data in a triple store and information in the the triple store can be queried using SPARQL.

The data extraction and transformation module included in PoolParty Semantic Integrator. UnifiedViews is an Extract-Transform-Load (ETL) tool, which can extract and convert data from various formats, such as CSV, JSON, XML, and relational tables as well as unstructured data such as PDF or text files into RDF for loading into PoolParty and connecting to a knowledge graph managed in PoolParty. It provides a graphical interface for creating pipelines with Data Processing Units (DPUs).

Stands for Uniform Resource Identifier, which identifies a resource on the internet or within a network of computers by specifying the protocol, web address, file location, and name of the object. URL (Uniform Resource Locator) is a kind of URI that also implies the means for retrieving the referenced resource. In RDF and RDF-based data structures (such as SKOS, RDF-S, and OWL) every resource, such as a concept or a class, is assigned a URI which makes it possible to unambigously reference within a knowledge organization system (e.g. a PoolParty project) or between linked vocabularies.

Stands for Vocabulary of Interlinked Datasets which is an RDF Schema vocabulary for expressing metadata about RDF datasets. It is intended as a bridge between the publishers and users of RDF data, with applications ranging from data discovery to cataloging and archiving of datasets. In PoolParty all metadata of a project along with some additional information is stored as a Vocabulary of Interlinked Datasets (VoID) graph.

Stands for World Wide Web Consortium which is the international standards organization for the World Wide Web, including the Semantic Web. It ensures compatibility for HTML, XML, RDF, OWL, SKOS and other coding schemes and their variations used for the web and also has published query languages such as SPARQL and SHACL.

A method used by a service offered by an electronic device to another electronic device, communicating with each other via the Internet. In PoolParty, Web Service Methods describe methods used in API calls.