**2. Why is XOL based on XML?**

*Ontological Analyses in Science, Technology and Informatics*

Accompanying with the development of Internet, more infrastructure, application, and knowledge base are generated. In normal knowledge supported systems, domain expert will first considerate the software environment and self-knowledge background and then choose the suitable knowledge scheme and ontology for the system. However, when it comes to the cross applications or large knowledgeassisted system, ontologies in system need to be reused. First, knowledge scheme in different systems may exist difference from the expert's personal cognize. Second, it offers several ontology languages for each system; thus, different ontology schemes can show in different formats, which make it hard to communicate in different applications. Third, the increasing demand of openness and the sharing lead of ontology could be exchange. Therefore, an ontology exchangeable protocol

or new ontology language supporting to exchange should be redefined.

ontology information across incompatible applications.

ontologies' use, exchange, negotiation, and cocreation.

Note the following XOL definitions:

<[slot-attribute]></[slot-attribute]>

class during the tag pair of <name></name>.

value's type of the attribute and the data restriction.

<name> [class-name] </name>

To realize the need of an evaluation on ontology in bioinformatics, several researchers on the US evaluation team developed a new specific ontology language—XOL [2]. By flexible XML expression and simplified protocol, XOL (xml-based ontology exchange language) is able to express and exchange different

XOL is an ontology language developing for exchange ontology in cross applications. It takes inspiration from OKBC (a protocol used for open knowledge base, see in Ref. [3]) and Ontolingua (another ontology used for reusing and editing ontology, see in Ref. [4]). Its syntax is based on human-readable and high compatible XML document. XOL can also respect as one effective intermedia language in

All of above XOL elements are pertained to all ontologies. Between the pair of <class></class> defines the basic information of this ontology, like the name of the

Pair of <slot></slot> will depict the attribute and restriction of the class, like

The last tag <individual> </individual> will give an instance of self-class or

With the human-readable and self-defined XML syntax, XOL can express the ontology in a concise way. However, it may also lead to the ontology inconformity

multiclass. It is not allowed to use the subclass as the individual element.

**1.2 Motivation**

**1.3 Definition**

**1.4 A simple example**

<class>

</class> <slot>

</slot> <individual> <name></name> <type></type> <…></…> </individual>

**36**

Generally speaking, each ontology language makes up for using syntax and language protocol. To realize the essence of XOL, we will show the different classifications of ontology languages based on the syntax and semantic rules.

According to the use of syntax, we can classify the ontology languages into three types as follows:

HTML format: Hypertext Markup Language (HTML) is the basic document mark of the current web. To extend the semantic character of the HTML, ontology language like SHOE offers an effective way to support semantic annotation by more extended webpage label.

XML format: Extensive Markup Language (XML) is a more human-readable and concise document for storing and defining different data. Ontology document made by XML format can easily locate by its hierarchical structure and semantic DTD tag.

RDF format: Resource Description Framework (RDF) is a new way to define ontology after XOL. It is a resource model always accompanied by a specific URI and extended specific XML-like label to depict the relation and knowledge model between the resources. It not only specifically and strictly expresses the data but also makes the alternation, merging, and inference possible.

According to language protocol of these languages, ontology language can divide into first-order predicate logic language, frame-based language, and concept-role restriction language.

First-order predicate logic language is the most accurate and original language in knowledge representation. The predicate formula is the formula formed by joining some predicates together with the predicate join symbol, like the largest formalized language Cyclo [5] and KIF [6].

Frame-based language is a language that includes the aforehand defining framework and simplified first-order logic language. Owing to excessive strict first-order predict logic and unreadable syntax, Ontolingua and frame logic are developed to remedy this defect.

Concept-role restriction language is an effort that most language currently adopts. This type of language offers a hierarchy way to represent the hyponymy by concept and the individual's signal. It reveals the relationship and value restrictions between different ontologies by role mark, like OML [7].

To note the difference between ontology languages in multiple syntax formats, we will give a detailed introduction for some ontology language with the technical developing route.

## **2.1 SHOE (HTML format)**

HTML had covered with a long history before the World Wide Web (WWW) appeared and is one of document standards of Standard Generalized Markup Language (SGML). SGML offers a high standard and complicated description about the document resource. As SGML is hard to learn, use, and realize, researcher put forward the HTML in 1989 after considering the computer's ability. HTML is the mere application of SGML in the WWW times. After few years of great development, HTML is widely known in the web document district. Semantic Web, as the next generation of WWW, is also the use of the HTML syntax in the ontology language. We called this simple HTML ontology language as SHOE.

SHOE is a specification that describes an extension to HTML, which provides a way to semantically describe important information about HTML or other web documents [1]. It offers a hierarchical classification mechanism for HTML documents and non-HTML documents or subsections of HTML documents. The intent of this specification is to make it possible for user agents, robots, and so on, to gather truly meaningful information about web pages and documents, enabling significantly better search mechanisms and knowledge gathering. Let us take the SHOE as an example; it can divide into two steps as follows:

#### **Define an ontology**

<Ontology "ontology-unique-name" version = "1.0" backward-compatiblewith = "version list">

#### **Use an existing XOL ontology**

<Ontology-extends "ontology-unique-name" version = "Version" backwardcompatible-with = "version list" URL = "location">

This is a simple way to define ontologies containing rules. Ontology simply means an ISA hierarchy of categories and a set of relations between these categories in this SHOE specification. Categories will also inherit relations defined in parent categories. However, this specification does not as yet define any other forms of relationships (transitive closures, inverses, negations, etc.) and use the complicated and human-unreadable Hypertext Markup Language as the basic syntax.

#### **2.2 KIF**

At the same time, first knowledge interchange format was proposed by American National Standard (dpANS). Though many ontology languages are still developed, researcher in Standard University begins to design a language for the intercommunication. Interchange of knowledge or ontology thought out disparate computer systems (different programmers, different languages, and other discrepancy in interknowledge sharing). KIF language is logically comprehensive with declarative semantics.

In addition to these essential features, KIF is designed to maximize the implementability and readability. KIF provides a declarative language for describing knowledge. As a pure specification language, KIF does not include commands for knowledge base query or manipulation.

#### **2.3 GFP**

GFP is first motivated by the hierarchic framework design of frame-based knowledge representation systems (FRSs) used at the Stanford Knowledge Systems Laboratory for accessing Cyc, KEE, and Epikit [8]. FRSs can contain all of the database systems and knowledge systems or other frame-like projects. It is complementally developed to support knowledge sharing. It specifies a new protocol, Generic Frame Protocol (GFP), for connecting knowledge bases (KBs) in FRSs. In more detail, it provides numbers of operations to formulate a general interface for all of the FRSs. Also, complementary tools were also produced to keep independent and general operation generation. GFP shows well compatibility between different languages, including Java, C (client implementation only), and Common Lisp. Thus, some format conversations of languages are also needed.

#### **2.4 OKBC**

After GFP coming out, OKBC, a new protocol called Open Knowledge Base Connectivity, has taken it up in more implicit knowledge model and knowledge

**39**

**Figure 1.**

*The structure of Ontolingua.*

*Ontology Language XOL Used for Cross-Application Communication*

operation. It first uses some open ontology systems, such as EcoCyc, GKB Editor, and Ontolingua projects. With 2 years of development, OKBC quickly used in

OKBC handles the knowledge in more implicit representation formalism, which we called OKBC Knowledge Model in later years. This model not only supports an object-oriented representation of knowledge but also can found the commonly knowledge structure from different KRSs. Therefore, it can serve as an implicit knowledge interlingua by its powerful character in knowledge for all of the systems

Ontolingua, which accompanies with different ontology languages breaking out, can serve as a basic framework to support open or domain knowledge sharing system. The syntax of Ontolingua definition is based on GFP. It is motivated by the need of Summer Ontology Project, a pilot study in which researchers from several groups and institutions met weekly to design ontology of terms used in modeling electromechanical devices. **Figure 1** depicts the structure of

*DOI: http://dx.doi.org/10.5772/intechopen.91385*

several ontology sharing projects.

using OKBC.

Ontolingua.

**2.5 Ontolingua**

*Ontology Language XOL Used for Cross-Application Communication DOI: http://dx.doi.org/10.5772/intechopen.91385*

operation. It first uses some open ontology systems, such as EcoCyc, GKB Editor, and Ontolingua projects. With 2 years of development, OKBC quickly used in several ontology sharing projects.

OKBC handles the knowledge in more implicit representation formalism, which we called OKBC Knowledge Model in later years. This model not only supports an object-oriented representation of knowledge but also can found the commonly knowledge structure from different KRSs. Therefore, it can serve as an implicit knowledge interlingua by its powerful character in knowledge for all of the systems using OKBC.
