JUMP - Java Metamodel Protocol

JUMP expresses database relationships in a RDF Resource Description Framework format, stored as a publicly readable file on a knowledge base server or database.
JUMP (an acronym for Java Metamodel Protocol) is a machine-readable ontology describing data objects in a database, their metadata, and their relations to other data objects. Anyone can use JUMP to describe structured data. JUMP allows organizations to describe the data residing in a database so that it can be presented as a web service.

JUMP's descriptive vocabulary is expressed using RDF Resource Description Framework and OWL Web Ontology Language. Computers may use these JUMP profiles to find data, for example, a natural language query “find all single nucleotide polymorphisms for TP53 gene in human and mouse orthologs”. In addition, JUMP leverages ontologies as higher-order langauges to map the corresponding database schema to a semantic standard. This is accomplished by defining the data and explicitly and linking the relationships between data. Each profile has a unique identifier (such as the database URL, or a URI, of the database server), where the data is stored.

The JUMP project, which defines and extends the vocabulary of a JUMP profile, was started in 2002 by Steve Perry and Kumar Sansar. It can be considered the first adaptation of Semantic Web standards for structured data, in that it combines RDF technology with SQL and Database metadata.

Example
The following JUMP profile (written in XML format) defines a data profile for a database containing patient records. JUMP provides a template for capturing many different types of metadata models - Conceptual schema models, Logical data model, Physical data model, relational models, rules models, etc. - that each provide specific metadata views. The JUMP framework delivers a structured meta-language to consolidate all the various models into a single view. This view captures the required integration detail as a data object for correctly cleansing and mapping the data. This level of detail typically resides in multiple models and must be consolidated as part of any integration project. JUMP provides a standardized way to document all required integration metadata in an object-oriented model to easily determine equivalencies across data domains.


  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#">
  
    
      
    
      
    
      
    
      
          
          
          
          
          
          
          
          
          
          
          
          
    
    
    
      
      
          full name of patient
          proper name
          string
          varchar(8)
          null
          fk.table.GLCTest
          fk.table.CaseNum

JUMP provides a lightweight metadata language that consolidates the required technical and business metadata associated with the data in a single view that can be stored as a reusable object.

JUMP Community
The JUMPER community sustains and supports the JUMP OWL schema. The team is completing the protocol component to allow JUMP files to be easily exchanged between systems for (semi) automated data integration. The goal of the JUMPER project is formalize a standards based specification around the schema and protocol.

JUMP is commonly used today as a proprietary method of capturing information about data residing in legacy systems. This information about the data is stored in a knowledge base. JUMPs modular format and consolidated views are an ideal tool for assembling legacy data in an easily reusable format. It is often used as an internal standard for documenting technical metadata across multiple projects. Based on need and requirement it has been modified and adapted across different organizations and industries.

The JUMPER Sourceforge project also provides open-source plugin scripts for building JUMP models within your favorite modeling tool.

Main Menu

See Also

Main Menu