Semantic Data Integration

If you have ever assembled a piece of furniture without instructions, you will understand how difficult it can be to create a finished product from many individual parts. Without the overarching context, i.e. knowing which piece of furniture you are supposed to build, successful assembly is almost impossible. With data, the situation is similar. Within their respective source system, instructions and target context are clearly defined. When it comes to the overall data and IT landscape of one or even several companies, however, things are less clear. There are many overarching issues and information requirements for which contexts either already exist or where they must be created first. Instructions in the form of prepared data models and definitions for information aggregation and data integration often do not exist – or they are only implicitly available in individual implementations. Semantic data integration allows all these aspects to be formally and explicitly described in a model. The context thus becomes tangible and can be optimized or extended depending on the application.

An illustrative example

Imagine an assembly kit for a bookcase, along with several related documents and information:

Instructions:

• Step-by-step instructions for the assembly
• Pictures of the individual steps

Parts list:

• A list of all required parts, with ID numbers and descriptions
• Required amount for each part (e.g. 4 screws, 2 side panels)

Tool list:

• A list of the tools that are required for the assembly (e.g. screwdriver, hammer)

Assembly:

To successfully assemble the bookcase, you will need to combine information from the instructions, parts list and tool list and understand how they are related to each other. For example:

• The instructions tell you that you need two side panels (part 1) and four screws (part 2) in the first step.
• The parts list gives you detailed information about what part 1 (side panel) and part 2 (screw) are and how many of each you have.
• The tool list informs you that you will need a screwdriver to fasten the screws.

It is by integrating and understanding this information that you can assemble the bookcase correctly. Semantic data integration helps you to understand the meaning and context of the individual pieces of information and combine them in a meaningful way.

Semantic data integration

With our methodical approach, we support you in analyzing benefits and identifying challenges and potentials for the development of an application context. This context is described semantically – typically modeled in ontologies – in order to record it in a formal, standardized and machine-readable way. Based on the developed context, we also analyze your existing database and information requirements from various systems for the overall context. Identified data is processed and cleaned according to the requirements of the application in order to integrate it using suitable integration mechanisms within an architecture that supports the application within the specified context. This way, we ensure data continuity and availability in the corresponding application context along the product life cycle through semantic data integration. Here we offer expertise and support in the development, implementation, application and maintenance of semantic data integration.

Our methodical approach ensures data continuity in the semantic integration. We help you define and successfully implement additional contexts for cross-system applications and analyses. Our approach ensures that the semantic architecture fits seamlessly into your existing system landscape and that both information quality and consistency are maintained for existing applications. Thanks to our many years of industry expertise, we can provide you with professional, technological, methodological and procedural support for the introduction and optimization of semantic integrations.