Report on RDFS of Knowledge Representation & Semantic Web Technologies

Resource Description Framework Schemas (RDFS) (Horrocks, et al., 2003; W3C, 2004) are the extensions of RDF that describe the classes and classification for the properties by using the RDF as knowledge representation language. The classes used in the RDFs enable to use of the resources and they are defined in classes and subclasses. The results provide fundamental processes and elements for the demonstration of ontologies. The frequent uses of elements in the RDF schemes include RDFS: range, RDFS: domain, RDFS: Type, RDFS: subclass, and RDFS sub-property. All the elements are mentioned below,

·         RDFS: the range of RDF: property: it is used to define the property ranges.

·         RDFS: domain of RDF: property: It specifies the domain of the property

·         RDFS: type: It defines that the instance demonstrates a type of class 

·         RDFS: subClassOf: It measures a complete class along with the subclass of some other class

·         RDFS: subPropertyOf: it is used to describe a complete and new property of a class and the sub-property of some other classes.

[D. Allemang and J. Hendler, Semantic Web for the Working Ontologist Effective Modeling in RDFS and OWL Second Edition publish a house.]

The examples used in the research illustrate the possibilities in which RDF can be used and it can be used in the real world things and relationships is between people and documents. In this relation, the classes can be classified in examples and consider the document, set of properties, and person. The described research illustrates the relation between the people and the documents. The documents are considered for the work and people work as agents.

3.4.1.5 Ontologies of Knowledge Representation & Semantic Web Technologies

Ontology is described as the spine that propels the Semantic Web architecture to provide the domains and machine-processable semantics. The process can be shared as well as it can assist the communication among people and other applications. Ontologies are mainly aimed to offer semantics, which creates a web of meaning. Ontologies work to assist the machines to progress and maintain the information as well as facilitate its sharing. (Buraga & Ciobanu, 2002). RIF (Rule Interchange Format) works to specify the XML format under different rules and the intermediate expressive power is further designed to match with RDF and OWL following what is explained by the RIF Working Group (W3C Recommendation, 2013a). Ontology OWL, Rules and Queries are described in more details in the subsequent sections as it is the main technologies used to model, implement and query the Smart- SSCCS framework.

3.4.1.6 Unified Logic of Knowledge Representation & Semantic Web Technologies

The logic is the layer in the semantic web stack that provides a reasoning engine designed to explain the semantic links and infer useful facts about the semantics.

3.4.1.7 Proof of Knowledge Representation & Semantic Web Technologies

This layer is used to check the legitimacy of specific statements and provides more Artificial intelligence (AI) mechanisms to make the web resource semantically reasonable to ensure reliable reasoning results.

3.4.1.8 Trust of Knowledge Representation & Semantic Web Technologies

Trust defines the last sheet in the semantic web stack, it tracks the trustworthiness of the information on the Web. Depending on policies and sources of information that are also available through the source, it can prevent other unwanted applications and the user is provided with access to all sources.

3.4.1.9 User Inference and Application

This layer is the baseline in which all applications and user interfaces must be satisfied.

3.4.1.10 Cryptography of Knowledge Representation & Semantic Web Technologies

The layers are digital signature and encryption. The initial conditions to start layer is from the first layer (URI, UNICODE) and then lead to the upper layer six (Proof). The digital signature can be demonstrated as a step headed for a web of trust. It is important to use the XML digital signature in the process and digital information can be marked. There are some components of XML syntax that can be applied in the procedure like Reference, SignedInfo, and DigestValue. XML Signatures are further used on the content of the specific resource and there is a way to all the resources that can be identified (W3C Recommendation, 2013b).

3.5 Ontology of Knowledge Representation & Semantic Web Technologies

The Semantic Web Technology (SWT) and its stack is widely known as the next

generation of the web. Its vision was to enhance web content with metadata; to enable human-machine readable content. This to process, share and interpret web content and deliver enhanced services. There are many successful implementations of general and domain-specific research that utilized SWT and its stack. For this research, we wish to focus on research that employed Web Ontology Language (OWL) Ontologies to enrich the web content semantically. The choice of SWT and Semantic Web Rule Language (SWRL) enabled OWL ontologies in specific was based on the flexibility the technology brings to software developers. Furthermore, OWL ontologies can play the main part in the SWT. Some of the benefits noticed while using SWT and its stack is then provided for a source under the shared and precisely defined domain-specific knowledge. Moreover, an ontology can consist of a conceptual schema of a domain; presented in a hierarchical explanation of essential ideas, together with the description of their properties (constraints) and enriched with the presence of domain-specific knowledge.

OWL ontologies also deliver added value of Semantic annotation that effective information retrieval. Hence, it enhances the internet with more capabilities of processing and accepting the semantics of information available on the internet. Consequently, it allows relevant information to be directly discovered.

“Ontology can be described as a file or document that formally defines all the relationships between the terms used in the documents. One of the typical types of ontology intended for the web shows the taxonomy and it provides a set of inference rules”. Taxonomies can include classifications and have a relationship defined through properties. The sub-hierarchal structure of taxonomies classes can provide explicit semantics of an environment. The ontologies can be enriched through the use of inference techniques. One advantage of using ontologies is that they enhance the functioning of the web in many ways. For example, they can improve accuracy based on concepts stored in the ontology(Berners-Lee, 2001).

3.5.1    Ontology definition of Knowledge Representation & Semantic Web Technologies

The word “ontology” has a different meaning in different communities when it is used. The most obvious differences are mentioned such as:

1.      The philosophical sense shows the well-known custom,

2.      The computational logic is developed in the current era and it is used in the knowledge engineering community. The computational sense is started from the informal description of (computational) ontologies that is “explicit specifications about the conceptualizations” (Guarino, Oberle, & Staab, 2009).

There is a different definition of ontologies in computer science. The following table summarises the definitions.

Table 2.1 – Definitions of an ontology (after Gomez-Perez et al., 2004)

 Table 3.x: definitions of ontologies in computer science adopted from(Asunción Gómez-Pérez Mariano, Fernández-López, Mariano, Oscar, 2015)

H. HERRE et al., General Formal Ontology (GFO): A Foundational Ontology Integrating Objects and Processes. Technical Report, 8, University of Leipzig, 2006

3.5.2    Different types of ontologies/ Ontology classification

Ontologies are classified into different dimensions. These range from certain generality levels and concepts to demonstrate the type of knowledge they model which can be associated to the domain or task (Guarino, 1998).: as can be seen in figure 3.x, the  arrows represent specialization relationships

• Top-level ontologies or Upper Ontology

Describes the overall perception and common-sense knowledge, for instance, the event, time, space, matter, action, and object. All the perceptions are independent of the specific problem or domain.

• Domain ontologies of Knowledge Representation & Semantic Web Technologies

It demonstrates the vocabulary associated with the entire generic domain, for example, physics otherwise medicine.

• Task ontologies

It describes vocabulary that relates to generic chores or actions such as analysis or trade.

• Application ontologies of Knowledge Representation & Semantic Web Technologies

It describes conceptualized ideas that depend on a specific domain and task. The application ontologies define specialized task ontologies and it corresponds to the domains and roles to perform certain activities. Van Heijst and the researchers proposed a method to classify the ontologies. The researchers considered two dimensions, amount and type of different structures, and the subject of conceptualization(Van H Eijst, Chreiber, & Ielinga, 1997)

1.      Amount and type of structure of the conceptualization: This dimension primarily looks at the level of granularity of the conceptualization and are subdivided into the following three classes:

•      Terminological Ontologies: Are lexicons that identify the terminology for ontologies, which are utilized to symbolize knowledge within the domain of communication. However, the semantics of the terms are not represented in this case. The ontology examples consider the medical field as a semantic system in the Unified Medical Language System (UMLS, Lindberg, Humphreys and McCray, 1993);

•      Information Ontologies: where databases (for example, database schemata) are specified through the record structure. Although this information provides a record of basic observations regarding instances of the database such as a student, they do not define the concepts that are instantiated by these instances such as learning style, ability, etc. A classic instance of such ontology is a structure designed for modeling the electronic medical records meant for patients (PEN & PAD model) (Rector, Nowlan, Kay, Goble, & Howkins, 1993)

·         Knowledge Modeling Ontologies: Special knowledge conceptualizations typically have a wealthier internal system than the knowledge modeling ontologies, and they frequently identified following the knowledge utilization that they described.

1.      The subject of the conceptualization: The knowledge type which is modeled within the ontologies described in this dimension. Four categories defined in this dimension as follow:

·         Application Ontologies of Knowledge Representation & Semantic Web Technologies

This category contains the entire fundamental concepts needed to model the requirement of knowledge for a particular application, and also could contain extensions of task-specific and method. The ontologies’ application commonly extends generic along with the domain knowledge by demonstrating both the method, as well as the components of task-specific taken from the generic and domain ontologies as the description below.

The knowledge that might be modeled in the libraries of ontology reused in the ontologies’ application by converting it for a further particular application. However, this ontologies’ application is considered to be reusable.

·         Domain Ontologies

These are specific conceptions of a specific domain. There is a differentiation made by knowledge engineers between domain knowledge (epistemic) and domain ontologies (ontological nature) such as:

-       Domain knowledge defines realistic situations

-       Domain ontologies more detail the application of limitations on the system along with the domain knowledge’s content

However, it is quite difficult to identify the differences between what is epistemological and ontological since they are slightly delicate.

·         Generic Ontologies of Knowledge Representation & Semantic Web Technologies

The generic ontologies’ concepts are similar to the domain ontology, which are identified as generic across various areas. Therefore, there is an unclear borderline between these two ontologies. Generic ontologies, however, have an intuitive useful and meaningful meant for constructing the building libraries since they match to the high-level of ontologies within the classification of Guarino. The generic ontologies usually describe concepts such as an event, state, process, as well as action.

·         Representation Ontologies of Knowledge Representation & Semantic Web Technologies

The underlying knowledge concepts explained by a formal representation. A representational framework provided here since they are quite neutral to the entities of the world. The domain and generic ontologies, however, are defined through the traditional set within these representation ontologies. The Frame Ontology which is utilized in Ontolingual is one of the examples in this category (Gruber, 1993).

Formality Degree of Knowledge Representation & Semantic Web Technologies

Ontologies could also have differed from the formality, whereas the terms’ meaning is articulated within the ontology (M. Uschold, 1996).The descriptions below that even though there could be the same knowledge articulated within the ontology, but the articulation way is different for them as follow:

·         Highly informal: The ontologies are articulated using a natural language. There might be vague in the definitions of the term due to there is an intrinsic uncertainty of the natural language itself.

·         Semi-informal: The ontologies are articulated here with a limited and structured natural language form. This would achieve the enhancement to the clarity, and at the same time, also reduce the ambiguity as well.

•      Semi-formal: The ontologies here are communicated within certain artificial languages which defined officially such as EasyOnto (Yadav, Murthy, Narula, Duhan, & Jain, 2018)

•      Rigorously formal: The ontologies here are using terms which are particularly described with formal semantics, theorems, as well as evidence of preferred properties like completeness and soundness.