Extracting knowledge from XML document repository: a semantic Web-based approach

XML plays an important role as the standard language for representing structured data for the traditional Web, and hence many Web-based knowledge management repositories store data and documents in XML. If semantics about the data are formally represented in an ontology, then it is possible to extract knowledge: This is done as ontology definitions and axioms are applied to XML data to automatically infer knowledge that is not explicitly represented in the repository. Ontologies also play a central role in realizing the burgeoning vision of the semantic Web, wherein data will be more sharable because their semantics will be represented in Web-accessible ontologies. In this paper, we demonstrate how an ontology can be used to extract knowledge from an exemplar XML repository of Shakespeare’s plays. We then implement an architecture for this ontology using de facto languages of the semantic Web including OWL and RuleML, thus preparing the ontology for use in data sharing. It has been predicted that the early adopters of the semantic Web will develop ontologies that leverage XML, provide intra-organizational value such as knowledge extraction capabilities that are irrespective of the semantic Web, and have the potential for inter-organizational data sharing over the semantic Web. The contribution of our proof-of-concept application, KROX, is that it serves as a blueprint for other ontology developers who believe that the growth of the semantic Web will unfold in this manner.

1. Greater detail to these ontology models is shown in [28].

2. → denotes parent-of.

3. Only some of the axioms need to answer this competency question is shown in this section. The remaining axioms as well a walk-through of the answering of the competency question is shown in the Appendix.

Authors and Affiliations

1. Schulich School of Business, York University, 4700 Keele St., Toronto, ON, Canada, M3J 1P3

   Henry M. Kim

2. Raj Soin College of Business, Wright State University, Dayton, OH, USA

   Arijit Sengupta

Corresponding author

Correspondence to Henry M. Kim.

Appendix

Defn-4. character_has_pseudonym(C,Ps)

A primitive description set describing one character can have both the character’s name and pseudonym used.

• character_has_pseudonym(C,Ps) =

• \({{\mathsf{\{ C, Ps|\exists Pd [primitive\_description\_set\_has\_character(Pd,C) \wedge}}}\)

• \({{\mathsf{primitive\_description\_set\_has\_psuedonym(Pd,Ps)\}}}}\)

Defn-5. \({{\mathbf{(a) related\_characters(C_{1}, Rn,Rp,C_{2})}}}\)

C₁ has a relationship, expressed as relation noun(Rn) + preposition(Rp), with C₂, if:

• C₁ and C₂ are characters in the same play.

• C₁ is explicitly stated as related to C₂ or C₂’s pseudonym, and

• C₁ is a character introduced individually, or is any of the characters in a group that has a relationship to C₂, and

  • \({{\mathbf{ related\_characters(C_{1}, Rn,Rp,C_{2}) = }}}\)

  • \({{\mathsf{\{ C_{1},Rn, Rp, C_{2}\vert}}}\)

  • \({{\mathsf{(\exists P (play\_has\_character(P,C_{1})\wedge play\_has\_character(P,C_{2})) \wedge}}}\)

  • \({{\mathsf{(\exists D(description\_has\_relationship(D,Rn,Rp,C_{2}) \vee}}}\)

  • \({{\mathsf{\exists Cr (description\_has\_relationship(D,Rn,Rp,Cr) \wedge}}}\)

  • \({{\mathsf{ character\_has\_pseudonym(C_{2},Cr)))\wedge}}}\)

  • \({{\mathsf{(primitive\_description\_set\_has\_character(D,C_{1}) \vee}}}\)

  • \({{\mathsf{\exists Pe \exists Pd (group\_has\_character\_description(D,Pe) \wedge}}}\)

  • \({{\mathsf{ character\_description\_has\_primitive\_description\_set(Pe,Pd)\wedge}}}\)

  • \({{\mathsf{ primitive\_description\_set\_has\_character(Pd,C_{1})))\}}}}\)

Defn-6. \({{\mathbf{(b) related\_characters(C_{1}, Rn,Rp,C_{2})}}}\)

C₁ has a relationship, expressed as relation noun(Rn) + preposition(Rp), with C₂ if:

• C₁ is a pseudonym for a character whose relationship with C₂ can be inferred, or

• C₂ is a pseudonym for a character whose relationship with C₁ can be inferred, or

• C₁ and C₂ are pseudonyms for characters whose relationship with each other can be inferred.

  • \({{\mathbf{ related\_characters(C_{1}, Rn,Rp,C_{2}) =}}}\)

  • \({{\mathsf{\{C_{1},Rn, Rp, C_{2}|\vert\exists C_{a}\exists C_{b}}}}\)

  • \({{\mathsf{(related\_characters(C_{1},Rn,Rp,C_{b})\wedge character\_has\_pseudonym (C_{b},C_{2}) \vee}}}\)

  • \({{\mathsf{(related\_characters(C_{a},Rn,Rp,C_{2}) \wedge character\_has\_pseudonym (C_{a},C_{1}) \vee}}}\)

  • \({{\mathsf{(related\_characters(C_{a},Rn,Rp,C_{b}) \wedge character\_has\_pseudonym (C_{a},C_{1}) \wedge}}}\)

  • \({{\mathsf{character\_has\_pseudonym(C_{b},C_{2})\}}}}\)

Defn-7. \({{\mathbf{ (a) may\_be\_related\_characters(C_{1}, Rn, Rp, C_{2})}}}\)

C₁ may have a relationship, expressed as relation noun(Rn) + preposition(Rp), with C₂, if:

• C₁ and C₂’s relationship (Rn + Rp) cannot be inferred for sure, and

• C₁ and C₂ are characters in the same play, and

• C₁ is explicitly stated as related to C₂’s qualifying title or location qualifier, and

• C₂ is a character introduced individually, or is any of the characters in a group, and

• C₁ is a character introduced individually, or is any of the characters in a group that has a relationship to C₂.

  • \({{\mathbf{ may\_be\_related\_characters(C_{1}, Rn,Rp,C_{2})=}}}\)

  • \({{\mathsf{ \{C_{1}, Rn, Rp, C_{2}\vert \neg related\_characters(C_{1},Rn,Rp,C_{2}) \wedge}}}\)

  • \({{\mathsf{(\exists P (play\_has\_character(P,C_{1}) \wedge play\_has\_character(P,C_{2})) \wedge}}}\)

  • \({{\mathsf{ (\exists C \exists D \exists D_{2}}}}\)

  • \({{\mathsf{(description\_has\_relationship(D,Rn,Rp,C) \wedge}}}\)

  • \({{\mathsf{ (description\_has\_qualifying\_title(D_{2},C)\vee description\_has\_location\_qualifier(D_{2},C)) \wedge}}}\)

  • \({{\mathsf{ (primitive\_description\_set\_has\_character(D_{2},C_{2}) \vee}}}\)

  • \({{\mathsf{ (\exists Pe_{2}\exists Pd_{2}(group\_has\_character\_description(D_{2}, Pe_{2}) \wedge}}}\)

  • \({{\mathsf{ character\_description\_has\_primitive\_description\_set(Pe_{2},Pd_{2})\wedge}}}\)

  • \({{\mathsf{ primitive\_description\_set\_has\_character(Pd_{2},C_{2}))) \wedge}}}\)

  • \({{\mathsf{ (primitive\_description\_set\_has\_character(D,C_{1}) \vee}}}\)

  • \({{\mathsf{ (\exists Pe \exists Pd (group\_has\_character\_description(D,Pe) \wedge }}}\)

  • \({{\mathsf{ character\_description\_has\_primitive\_description\_set(Pe,Pd)\wedge}}}\)

  • \({{\mathsf{ primitive\_description\_set\_has\_character(Pd,C_{1}))) \}}}}\)

Defn-8. \({{\mathbf{has\_son(C_{1}, C_{2}) = }}}\)

\({{\mathsf{ \{C_{1}C_{2} \vert related\_characters(C_{2},`son\hbox{'},`of\hbox{'},C_{1}) \vee related\_characters(C_{2}, `son\hbox{'}, `to\hbox{'},C_{1}) \} }}}\)

Defn-9. \({{\mathbf{has\_father(C_{1},C_{2}) =}}}\)

\({{\mathsf{ \{C_{1}C_{2}\vert related\_characters(C_{2},`father\hbox{'}, `of\hbox{'},C_{1}) \vee related\_characters(C_{2},`father\hbox{'}, `to\hbox{'},C_{1}) \}}}}\)

Defn-10. male(C) =

\({{\mathsf{\{C| \exists C_{1} has\_son(C_{1},C) \vee has\_father(C_{1},C) \}}}}\)

Defn-11. \({{\mathbf{has\_child(C_{1},C_{2}) =}}}\)

\({{\mathsf{ \{C_{1}, C_{2} \vert has\_son(C_{1},C_{2})\vee has\_father(C_{2},C_{1}) \}}}}\)

Obviously, many such relationship terms can be defined, e.g. daughter of, mother of, an additional definition of parent of, uncle of, etc. Also possible familial relationships can be defined using may_be_related_characters.

Definitions for answering CQ-2 and CQ-3 are straightforward, so are not presented. The predicate play_has_character has been defined, so CQ-4 can be answered.

In the next section, these axioms are applied to answer competency questions (Fig. 11).

Excerpt from XML document of ‘Romeo and Juliet’ [7]

Full size image

1.1 Demonstration of competency

Following are some primitive terms.

1.2 Relevant primitive term instances

With that, the following competency question can be answered

CQ-1. Which character is the son of the Montague character?

Answering CQ-1

Reprints and permissions