Semantic Web

Download

.ppt.doc


ABSTRACT

We define semantic technology as a software technology that allows the meaning of and associations between information to be known and processed at execution time. For a semantic technology to be truly at work within a system there must be a knowledge model of some part of the world that is used by one or more applications at execution time.How is it distinguished from more conventional applications?Semantic technologies represent meaning through connectivity. The meaning of terms, or concepts, in the model is established by the way they connect to each other.

  A semantic model expresses multiple viewpoints.Semantic models represent knowledge about the world in which the system operates. Several interconnected models could be used to represent different aspects of the knowledge. The models are consultable (accessible) by applications at run time.

  A semantic application uses knowledge models in an essential way as part of its operation. Use of a model is often referred to as "reasoning over the model". Reasoning can range from a very simple process of graph search to intricate inferencing over the model.

  Semantic applications are thin because they work with “smart” data. All the business rules logic is held in the models shared across applications. Simplest  form of a semantic model, a taxonomy. The model describes government concepts that are part of Federal Enterprise Architecture (FEA). In a taxonomy connections between terms exist, but are not named. Therefore, the structure itself becomes a way to identify the nature of relationships. Taxonomies are hierarchies that establish “parent-child” relationship between its concepts.

                                           Introduction

 

Semantic Technology

What is Semantic Technology?

We define semantic technology as a software technology that allows the meaning of and associations between information to be known and processed at execution time. For a semantic technology to be truly at work within a system there must be a knowledge model of some part of the world that is used by one or more applications at execution time.How is it distinguished from more conventional applications?

 Semantic technologies represent meaning through connectivity. The meaning of terms, or concepts, in the model is established by the way they connect to each other.

 A semantic model expresses multiple viewpoints.

 Semantic models represent knowledge about the world in which the system operates. Several interconnected models could be used to represent different aspects of the knowledge. The models are consultable (accessible) by applications at run time.

 A semantic application uses knowledge models in an essential way as part of its operation. Use of a model is often referred to as "reasoning over the model". Reasoning can range from a very simple process of graph search to intricate inferencing over the model.

 Semantic applications are thin because they work with “smart” data. All the business rules logic is held in the models shared across applications. Simplest  form of a semantic model, a taxonomy. The model describes government concepts that are part of Federal Enterprise Architecture (FEA). In a taxonomy connections between terms exist, but are not named. Therefore, the structure itself becomes a way to identify the nature of relationships. Taxonomies are hierarchies that establish “parent-child” relationship between its concepts.

THE GOAL OF THE SEMANTIC WEB

  The Semantic Web and Semantic Web technologies offer us a new approach to managing information and processes, the fundamental principle of which is the creation and use of semantic metadata.For information, metadata can exist at two levels. On the one hand, they may describe a document, for example a web page, or part of a document, for example a paragraph.

  On the other hand, they may describe entities within the document, for example a person or company. In any case, the important thing is that the metadata is semantic, that is it tells us about the content of a document (e.g. its subject matter, or relationship to other documents) or about an entity within the document. This contrasts with the metadata on today’s Web, encoded in HTML, which purely describes the format in which the information should be presented: using HTML, you can specify that a given string should be displayed in bold, red font but you cannot specify that the string denotes a product price, or an author’s name, and so on.

  There are a number of additional services which this metadata can enable . In the first place, we can organise and find information based on meaning, not just text. Using semantics our systems can understand where words or phrases are equivalent. When searching for ‘George W Bush’ we may be provided with an equally valid document referring to ‘The President of the U.S.A.’. Conversely they can distinguish where the same word is used with different meanings. When searching for references to ‘Jaguar’ in the context of the motor industry, the system can disregard references to big cats.

  When little can be found on the subject of a search, the system can try instead to locate information on a semantically related subject. Using semantics we can improve the way information is presented. At its simplest, instead of a search providing a linear list of results, the results can be clustered by meaning. So that a search for ‘Jaguar’ can provide documents clustered according to whether they are about cars, big cats, or different subjects all together. However, we can go further than this by using semantics to merge information from all relevant documents, removing redundancy, and summarising where appropriate. Relationships between key entities in the documents can be represented, perhaps visually.

  Supporting all this is the ability to reason, that is to draw inferences from the existing knowledge to create new knowledge. The use of semantic metadata is also crucial to integrating information from heterogeneous sources, whether within one organisation or across organisations. Typically, different schemas are used to describe and classify information, and different terminologies are used within the information. By creating mappings between, for example, the different schemas, it is possible to create a unified view and to achieve interoperability between the processes which use the information.

  Semantic descriptions can also be applied to processes, for example represented as web services. When the function of a web service can be described semantically, then that web service can be discovered more easily. When existing web services are provided with metadata describing their function and context, then new web services can be automatically composed by the combination of these existing web services. The use of such semantic descriptions is likely to be essential to achieve large-scale implementations of an SOA.

No comments: