- Email: [email protected]
Ontologies for intellectual property rights protection
Expert Systems with Applications 39 (2012) 1388–1400
Contents lists available at SciVerse ScienceDirect
Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa
Ontologies for intellectual property rights protection Zhang X.M. a,⇑, Liu Q. a, Wang H.Q. b a b
School of Information Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan, Hubei Province, PR China Department of Information Systems, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
a r t i c l e
i n f o
Keywords: Ontology Intellectual property rights (IPR) Copyright protection Ontology Web Language (OWL)
a b s t r a c t Pirating various forms of intellectual property (IP) causes great economic loss to intellectual property rights (IPR) holders. IPR protection is becoming a key issue in our highly networked world. In order to further deepen our understanding of how to protect IPR and enhance information interchange and knowledge sharing among related entities, ontologies for IPR protection are proposed. This study contains three parts, which are developed to deal with different perspectives in this domain. The first part presents a static ontology, i.e. a hierarchy framework for the domain language, including primarily classes of participants, classes of IP works, classes of activities, and relations between these classes. In the second part, a dynamic ontology is shown to illustrate the IPR protection process. Thirdly, a causal map is used to demonstrate how classes of IPR protection methodologies are causally related with classes of IP piracy methodologies. Finally, the case of Tomato Garden is offered to demonstrate how the proposed ontologies are used in the real world. In respect of the ontology, it is first helpful to gain a comprehensive understanding of domain knowledge of IPR protection; second, IPR protection systems’ design and development in this domain are facilitated and supported by these ontologies; third, the proposed ontologies are united in the Ontology Web Language (OWL) and the OWL rules languages framework, both of which are machine readable. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction The World Intellectual Property Organization (WIPO), defines intellectual property (IP) as creations of the mind: inventions, literary and artistic works, and symbols, names, images, and designs used in commerce. IP is divided into two categories: Industrial property, which includes inventions (patents), trademarks, industrial designs, and geographic indications of source; and Copyright, which includes literary and artistic works such as novels, poems and plays, films, musical works, artistic works such as drawings, paintings, photographs and sculptures, and architectural designs (WIPO, 2004). IP allows people to own their creativity and innovation in the same way that they can own physical property. The owner of IP can control and be rewarded for its use, and this encourages further innovation and creativity to the benefit of everyone. Often, more than one of the above protection types may apply to the same creation. Innovation in information technologies and network communications offers people a great opportunity for the widespread and efficient utilization of IP works through various channels. As well as enjoying the convenience of worldwide information sharing, however, the entire society is faced with the issue of violation of ⇑ Corresponding author. Tel.: +86 2787298927; fax: +86 2787665520. E-mail address: [email protected] (X.M. Zhang). 0957-4174/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2011.08.021
IPR. IPR violation overlaps with issues of commercial domain, legal domain and technical domain – piracy of software, audio, database, books, reverse engineering of marketed hardware as well as theft of sensitive commercial designs by competing corporate entities. The piracy of IP works is a major form of IP violation. The International Intellectual Property Alliance (IIPA) estimated the annual loss of revenue in the US business software industry due to piracy at US$14273 million, and in the record and music industries at US$1486.9 million, for the financial year of 2009, as reported on 18 February 2010 (IIPA, 2010). It is also worth noting that a large portion of Internet bandwidth (approximately 30%) is consumed by users exchanging illegal copies of digital media (mainly video). It is certain that there will always be people with enough motivation to illegally use IP works by circumventing protection mechanisms (Vassiliadis & Fotopoulos, 2007). It is the goal of our paper to propose ontologies that illustrate the domain knowledge about IPR protection. The paper is comprised of three parts. The first part, which is represented using the description logic variant of the Web Ontology Language (OWL DL), provides a static ontology, i.e. a hierarchy framework for the domain language, including primarily classes of participants, classes of IP works, classes of activities, and relations between these classes. It constitutes a specification of the domain-specific concepts of classes, entities, properties, and activities as a set of relationships that exist among these vocabulary terms. In the second part, a dynamic ontology is presented to illustrate
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
the IPR protection process. Thirdly, a causal map is used to demonstrate how classes of IPR protection methodology are causally related with classes of IP pirate methodology, which can be written as rules using the OWL rules language (Horrocks, Patel-Schneider, Bechhofer, & Tsarkov, 2005). The ontology in the paper provides not only a formal description of objects in the domain knowledge and shared terminology, but also a formal basis for reasoning domain knowledge. Thus on the basis of the ontologies, it is first helpful to gain a comprehensive understanding of the domain knowledge of IPR protection; second, IPR protection systems’ design and development in this domain are facilitated and supported by this ontology. The proposed ontologies are then united in the OWL and the OWL rules languages framework, both of which are machine readable, part of which is shown in Appendix A. The rest of the paper is organized as follows: Section 2 illustrates the related techniques, i.e. ontology, OWL DL, and OWL rules language; the details of static and dynamic ontologies are presented in Section 3. The causal map in the domain will be proposed in Section 4. In Section 5, the Tomato Garden case is analyzed using the proposed ontologies. Finally, conclusions are presented in Section 6.
2. Background 2.1. Related works In order to solve the problem of IPR violation, many digital rights management (DRM) systems are proposed in literature. Some significant references include: Camp (2003), illustrating first principles of copyright for DRM design; Torres, Serrao, Dias, and Delgado (2008), offering an analysis of the various methods for implementing interoperable digital rights management platforms; Jamkhedkar and Heileman (2008), analyzing the problems with current DRM environments and proposing an open layered framework for the development of DRM systems; Thomas, Emmanuel, Subramanyam, and Kankanhalli (2009) proposing a joint digital watermarking scheme using the Chinese remainder theorem for the multiparty multilevel DRM architecture; and Lee et al. (2005), for designing a contents distribution framework that supports transparent distribution of digital contents on the Internet as well as the copyright protection of participants in the contents distribution value chain. A useful analysis of DRM business models, standards, and core technologies can be found in Koenen, Lacy, MacKay, and Mitchell (2004), Ku and Chi (2004), Cohen (2003), and Felten (2003). The increasing use of mobile devices has also initiated research efforts for mobile DRMs (MDRMs); technological challenges in this area differ from classic DRM and include mobile device limitations, bandwidth, usability, among others (Chen, 2008; Lee, 2007). Many companies, organizations and administration-funded projects provide solutions for the implementation of DRM systems, which has given rise to a boom in commercial DRM systems, such as the Adobe e-book for pdf documents, the IBM cryptolope (Kaplan, 1996), the ambitious DigiBox technology by InterTRust (Kohl, Lotspiech, & Kaplan, 1997), Microsoft’s Windows Media Player for audio/video, and Digimarc’s family of products for video/audio and still images, to name a few. Nevertheless, although these solutions have several aspects in common, they are incompatible in terms of architecture and system components. Moreover, efforts that facilitate copyright management in closed domains experience great difficulty when they are forced to interoperate in an open domain like the World Wide Web. In order to facilitate interoperation and automation, DRM systems can be enriched with domain formalizations. There are many other initiatives to build an IPR framework for the Internet-wide management of IP works; for instance, MPEG21 (MPEG, 2002) or, in the W3C (World Wide Web Consortium)
1389
initiatives framework, the ODRL (Open Digital Rights Language) proposal (Iannella, 2002). Most of these initiatives have one thing in common: they work at the syntactic level. Their approach is to formalize XML DTDs (Document Type Definition) and Schemas that define a rights expression language (REL), such as the MPEG-21 REL (MPEG, 2003). In some cases, the semantics of these languages, i.e. the meaning of the expressions, are also provided but formalized separately as rights data dictionaries (RDD); for instance, MPEG21 RDD (MPEG, 2003). Rights dictionaries define terms in natural language, solely for human consumption; for this reason it is not easy to automatically process those terms. However, the syntactic approach does not scale well in really wide and open domains like the Internet. The automatic processing of a huge amount of metadata coming from many different sources requires machine understandable semantics. The syntax is inadequate when unforeseen expressions are encountered, which is where semantics can assist with their interpretation to achieve interoperation. Delgado, Gallego, Llorente, and García (2003), proposed an ontology for DRM which concentrates only on some aspects of IPR protection, in particular, the concepts of IPR Agreement, including Contract and License, and some specific rights, such as exploitation or moral rights. A copyright ontology is implemented in García and Gil (2006), using OWL DL. This approach facilitates the implementation of efficient usages against license checking, which is reduced to description logics classification. In order to support the whole copyrighted content value chain across enterprise or business niches boundaries, García and Gil (2010) provides a framework that accommodates copyright law and a rich creation model in order to cope with all the creation life cycle stages. The dynamic aspects of value chains are modeled using a hybrid approach that combines ontology-based and rule-based mechanisms. These ontologies were constructed from a rights management point of view; in particular, the concepts of Contract and License. Nevertheless, there is a need for a more comprehensive ontology to illustrate the IPR protection issue.
2.2. Ontology As a branch of philosophy, ontology is the study of the kinds of things that exist. It is often said that ontologies ‘carve the world at its joints’ (an expression commonly attributed to Plato). In the artificial intelligence (AI) community, the term ontology is used to refer to a set of representation vocabulary, and, more precisely, it is the conceptualizations in a domain that the terms in the vocabulary are intended to capture (Chandrasekaran, Josephson, & Benjamins, 1999). According to Sowa, knowledge representation can be defined as the application of logic and ontology to the task of constructing computable models of a domain (Sowa, 2000). Concerning the nature and relations of being, ontologies can be used as critical components in knowledge management and have huge potential to improve information organization, management, and understanding. In order to further deepen our understanding of IPR protection issues and enhance information and knowledge sharing among related entities, ontologies for IPR protection issues are proposed in this study. Jurisica, Mylopoulos, and Yu (2004), have classified ontologies into four broad categories: static, dynamic, intentional, and social. Two of these ontologies will be considered here: those that deal with static and dynamic aspects of IPR protection issues. Static ontologies represent the static aspect and define the basic concepts of IPR protection issues, while dynamic ontologies deal with knowledge regarding the IPR protection process. This paper will also discuss the implementation level at which OWL and OWL rules languages are used to represent these ontologies in a machine-readable form.
1390
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
2.3. OWL and OWL rules languages OWL is a good representation for ontology and was developed by the world standards body, W3C (Smith, Welty, & McGuiness, 2004). It is designed for use by applications that need to process the content of information rather than simply presenting information. It allows more vocabulary terms to be used for describing properties and classes. The main advantages of OWL are efficient reasoning support, sufficient expressive power, and convenient expression. OWL offers three sublanguages: OWL Lite, OWL DL, and OWL Full. OWL DL supports those users who want maximum expressiveness while retaining computational completeness and decidability. In our study, OWL DL is suitable for representing the various types of classes and relationships in the ontology. Although the OWL Web Ontology Language has considerable expressive power, it also has expressive limitations, particularly with respect to what can be said about properties. In order to overcome these limitations, Horrocks et al. (2004) proposed the Semantic Web Rules Languages (SWRL). SWRL extends OWL with a simple form of Horn-style rules (Horrocks et al., 2004). The proposed rules are of an implication form between an antecedent (body) and consequent (head). Both the antecedent and consequent of a rule consist of zero or more atoms. The informal meaning of a rule can be read as: whenever the conditions specified in the antecedent hold, then the conditions specified in the consequent will also hold. Atoms in rules can be of the form C(x), P(x, y), Q(x, z), same as (x, y), or different (x, y), where C is an OWL DL description, P is an OWL DL individual-valued Property, Q is an OWL DL data valued Property, x and y are either variables or OWL individuals, and z is either a variable or an OWL data value. 3. Ontologies for IPR protection In this section, ontologies for IPR protection will be given, and will include static and dynamic ontologies. 3.1. Static ontology As noted above, the static ontology represents the static aspect of the IPR protection issue and defines the basic concepts for that issue, e.g. IP works, information technology, corporate and so forth. As analyzed in Section 1, the IPR domain includes various kinds of entities, and there are complex relationships among them. Fig. 1(a) shows the top-level ontology of IPR protection, through which general knowledge is given. Three classes are shown in the top level of the static ontology: the IP works class, the participants class, and the activities class. The IP works class comprises various forms of IP that are the most likely to be violated, namely, the IP distribution channel and IP lifecycle. The participants class is used to represent the roles involved in IPR protection issues. According to the interest gained from IP Works, the roles are classified into three subclasses: pirate, anti-pirate and neutrality. Each of them has individual, corporate and other types. The activities class describes the actions taken by the participants. It mainly involves an IPR protection methodology subclass, which provides the essential approaches supporting IPR protection, and the piracy methodology subclass, which presents the most commonly used piracy approaches. These three classes have complex relationships; in addition, their subclasses also have intricate relationships with each other. The details of these relationships at the class level will be discussed in subsequent sections. Fig. 1(b) shows the subclass of the form of IP works in detail. The IP works class is made up of various forms of IP that are most likely to be violated. The act of piracy is most likely to be committed against the three main subclasses of IP works: multimedia,
software and design. The most common types of multimedia include text, image, audio, video, and similar forms. Operating systems and applications are the two main software types. Designs such as blueprints, circuit design, database, medicinal compounds, cosmetics, etc., make up another form that is highly likely to be violated. Fig. 1(c) illustrates the primary distribution channels of IP works. Traditionally, the physical distribution channels are CD, DVD, Book, or some other fixation medium subject to copyright. With the extensive use of networks, the online shop has become a very popular distribution channel for IP works. Another distribution channel through networks is that of super distribution by P2P file sharing, email, instant messaging, and so on. Fig. 1(d) describes the lifecycle of IP works. There are four main stages in the lifecycle: the creation stage, assertion stage, distribution stage and usage stage. IPR protection methodology class is shown in Fig. 1(e). Its subclasses are all possible techniques to protect IPR or prevent piracy. The most commonly used techniques are encryption algorithms, such as DES, AES, RSA, Elliptic Curve and so forth; metadata; digital certificates, such as X509; digital signature, such as RSA signature or hash signature; digital watermarking – a promising solution to copyright problems – such as copyright watermark, authentication watermark, and usage control watermark. Finally, the piracy methodology class is presented in Fig. 1(f). The most commonly used pirate methods are copying, downloading, reuse, dishonest claims, superdistribution, rerecord, and crack.
3.2. Relationships between participants and activities In Fig. 2, the Participants class is classified into three types: Pirate, Anti-Pirate and Neutrality. These participants represent the commercial entities in the real world. Piracy occurs in one of two ways: naive or malicious (Reisman & Heights, 2006). The way in which it occurs is determined by the motivations of commercial entities. The pirate class is composed of corporate entities, institutions, individuals A company may intend to pirate the IP of its opponent in order to gain greater profit and outmaneuver its commercial competitors. An individual may crack software and share it with friends. An institution may install unauthorized software on its computers. Whether the piracy takes place with naive or malicious motivations, it results in loss of profit for the IPR holders, who mainly fall into the anti-pirate category. The anti-pirate category consists of government, institutions, corporate entities, and individuals. All the members of this group try to prevent or reduce piracy and, if necessary, to punish the pirate. Neutrality exists as a third category, which is most likely to be made up of potential customers of certain IP works. Once conditions change, a customer may turn to either of the other two categories. For example, Alice wants to buy a CD from an online shop. If she finds the CD can be obtained free on a website, then she is most likely to change her mind and download the CD from the website. The left side of this figure shows the classes of Participants. The right side shows the classes of activities. The dashed open arrows link two different classes, and denote their Objectproperties. For example, if, at the class level, the arrow labeled ‘‘adopted by’’ links a Participants class and an Activities class, it means that the objects in the Activities class are adopted by the objects in the Participants class. Meanwhile, the arrow labeled ‘‘Invent’’ means that the objects in the Participants class invent the objects in the Activities class. For example, at the instance level, a copyright watermark is embedded into original data to identify its ownership by the author, John. The distributor, ADS Ltd, can adopt an X.509 Certificate to protect the content in the distribution stage. The pirate company, HH, produces a pirate CD by using Combo CD writer.
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
1391
Fig. 1. Class of static ontology. (a) Class of static ontology. (b) IP works class. (c) IP distribution channel class. (d) IP lifecycle class. (e) IPR protection methodology class. (f) Piracy methodology class.
1392
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 2. Relationships between participants and activities.
The famous encryption algorithm, RSA was invented by Ron Rivest, AdiShamir, and Leonard Adleman. 3.3. Relationships between Participants and IP Works As shown in Fig. 3, the left side in the figure is the classes of IP works. IP works in different forms are distributed through different IP distribution channels. Each IP work will go through an IP lifecycle, including a creation stage, assertion stage, distribution stage and usage stage. The right side of this figure is the classes of Participants. As shown at the class level, the arrow labeled ‘‘take activities around’’ links the Participants class and IP works class, which means that the objects in the Participants class take activities around the objects in the IP works class. Meanwhile, the arrow labeled ‘‘have effect on’’ means that the objects in the IP works class have an effect on the objects in the Participants class. For example, at the instance level, a potential customer may pirate anti-virus software distributed by P2P file sharing. A pirate company intends to pirate the circuit design through the manufacturer. A creator is active in the creation stage of the IP lifecycle and corporate distributor is active in the distribution stage of the IP lifecycle. 3.4. Dynamic ontology of IPR protection Dynamic ontologies describe changing aspects of the world. Typical primitive concepts include state, state transition and process. The dynamic view of IPR protection allows the construction of IPR protection models for use in electronic commerce. This includes the possibility of defining events in the IP works lifecycle. In this section, the dynamic ontology deals with knowledge of the IPR protection process and a UML state diagram, shown in Fig. 4, is adopted to depict it.
In order to provide a detailed description of the IPR protection process, Fig. 4 has several parts. Fig. 4(a) shows that the general IP protection process consists of five parts: the creation process, assertion process, usage process and monitoring process. Fig. 4(b1) shows the IPR protection process in the creation stage. At the beginning of the process, a creator creates content (creation stage). If the creator wants to protect his creation from piracy, he can apply for IPR protection methodology. In this stage, the creation data is saved on a computer. As a result, different IPR protection methodologies are selected to protect the computer and the creation data. Here for instance, access control is used to protect the computer from unauthorized access, and encryption is used to protect the creation data. After this deployment, the creation data is in a protected state. Fig. 4(b2) shows the pirate process in the creation stage. A pirate who wants to steal the creation data in this stage will apply for appropriate pirate methodologies. In the situation that creation data saved on a computer, the pirate firstly illegal accesses the computer. If the computer is protected by an access control policy, it will check whether the pirate access rights. Once the rights check confirms this is an illegal access, the pirate attempt is prevented. In the event that the rights check confirms an authorized access, the pirate will enter the operating system and apply other pirate methodologies to obtain the creation data. If the creation data is encrypted by the creator, the pirate will try to decrypt it. Only if the decryption is successful can the pirate obtain the creation data by copying or downloading. Fig. 4(c1) illustrates the IPR protection process in the assertion stage. After completion of the creation, the creator should assert content when it is first created (or is reused and extended with appropriate rights). Firstly, the creator has to submit the content and metadata to the authority for request. The authority applies for appropriate protection methodologies, i.e. digital fingerprinting
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
1393
Fig. 3. Relationships between participants and IP works.
to identify the content, and to ensure that the same content is not already registered. If the same content exists, the assertion fails. Otherwise, the authority confirms this is novel content and the creation is validated. The next step is rights assertion. The creator makes the rights request to the authority and the latter applies for IPR protection methodologies, i.e. the generation of a digital watermark/certificate, which is distributed to the creator to identify him or her as the rights holder. Fig. 4(c2) describes the pirate process in the assertion stage. The most likely scenario is that the pirate has obtained a copy of the creation data at an early stage and then dishonestly claims ownership of this creation to the authority. Through checking digital fingerprinting/watermarking, the authority identifies whether the content is a novel creation, in Fig. 4(c1). If the real creator has successfully asserted rights, the pirate will be prevented from proceeding. The IPR protection process in the distribution stage is shown in Fig. 4(d1). The creator can make a request to the authority to transfer the IPR. The authority applies for IPR protection methodologies, generating digital certificate/watermarking and distributing it to the rights holder and distributor. Meanwhile, the creator creates a business model for the IP works, which is codified as rules. The content and codified rules are packaged with encryption/certificate/watermarking and prepared for distribution. The distributor is responsible for managing and enabling the trade of the content. In order to access the content, the customer has to complete payment through the authority. On receiving payment from the customer, the authority confirms this to the distributor, who generates the digital certificate/watermarking which carries the usage rights and distributes it to the customer. The pirate process in the distribution stage is shown in Fig. 4(d2). At the start of the process, the pirate makes an application to the distributor to view the content and obtains the watermarked content/temporal certificate. The pirate then applies pirate methodologies to complete the theft by, for instance, rerecording the content or keeping an illegal copy of the content. Final-
ly, the pirate obtains watermarked content, which exposes him/her to high risk. Fig. 4(e1) illustrates the IPR protection process in the usage stage. In the initial stage, the customer obtains encrypted/watermarked content and the end-user player in the distribution stage. Before the content is used by the customer, the IPR protection methodologies are carried out to check whether the customer holds the legal usage rights. If the check fails, usage is forbidden. Fig. 4(e2) illustrates the pirate process in the usage stage. This type of piracy is most likely to impact a customer who has purchased authorized content. The pirate applies pirate methodologies to copy/download/crack the customer’s authorized content and obtains a copy of the content. The pirate then reuses the content, packages and sells the content, or shares the content with others through superdistribution, or similar means. The IPR protection process in the monitoring stage is shown in Fig. 4(f). After selling the content to the customer, the distributor will monitor usage through IPR protection methodologies. The most commonly used methodology is agent technique. The crawling agent searches suspicious content in the networks, checks whether there is a digital watermark/fingerprint/certificate in the content, makes comparison with the record database and judges whether the content has been pirated or is unauthorized. A distributor who has uncovered illegal usage has the right to bring legal action against the pirate.
4. Causal maps in the domain 4.1. Causal maps A causal map offers to model interrelationships among a variety of concepts; it can be employed to cope with complicated problems for which analytical techniques are inadequate. It is one of five generic types of cognitive map, and is still the most popular mapping method (Chaib-draa, 2002). When a causal map is
1394
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 4. Dynamic ontology for IPR protection. (a) IPR protection process for IP works. (b1) IPR protection process in creation stage. (b2) Pirate process in creation stage. (c1) IPR protection process in assertion stage. (c2) Pirate process in assertion stage. (d1) IPR protection process in distribution stage. (d2) Pirate process in distribution stage. (e1) IPR protection process in usage stage. (e2) Pirate process in usage stage. (f) IPR protection process in monitoring stage.
pictured in graph form, it is relatively easy to see how concepts and causal relationships are related to each other and to see the overall causal relationships of one concept with another. A causal map M is a pair, M = (C, E), where C is a set of nodes, or concepts, and E is a set of signed edges (Wellman, 1994). E = {(c, c0 , d): c, c0 e C, there is an edge of sign d from c to c0 }. The possible signs are usually ‘‘+’’, ‘‘’’, ‘‘0’’ and ‘‘?’’. They respectively denote ‘‘c has increasingly effect on c0 ’’, ‘‘c has decreasingly
effect on c0 ’’, ‘‘c has no effect on c0 ’’, and ‘‘c has ambiguous effect on c0 ’’. Usually, there are two types of concept events in causal maps: existential and directional. The directional event is used in this paper. For c 2 C, c is a directional event. Positive change, denoted by c(+), represents it is known that c is rising.
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 4 (continued)
1395
1396
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Zero change, denoted by c(0), represents it is known that c is unchanged. Negative change, denoted by c(), represents it is known that c is decreasing. d
þ
(c, c , d)((c, c , d) e E) can be written as rule: c ! c . c ! c means that if c is ‘‘+’’(‘‘’’), then c0 will hold as ‘‘+’’(‘‘’’). c ! c0 means that 0 if c is ‘‘+’’(‘‘’’), then c0 will hold as ‘‘’’(‘‘+’’). And c ! c0 means that c ? has no effect on c0 . c ! c0 means that c has ambiguous effect on c0 . A causal map graph is a set of classes connected by arrows, with the direction of the arrow indicating the direction of influence or causality. The general definition of derived relationships between in an acyclic causal map is simply the sum of the signs for all paths between them. Let Pa,b be the set of paths in M from a to b, then the sign of derived causal relation from a to b is: p2Pa;b ððc;c0 ;dÞ2p dÞ. 0
0
0
0
Table 1 The annotations used in Fig. 5. IPR protection method AC E MD CW AW UCW DC DS DF SD
Access control Encryption Metadata Copyright watermark Authentication watermark Usage control watermark Digital certificate Digital signature Digital fingerprinting Specific device
Table 2 Dataproperty of variables of IPR protection method.
4.2. Causal maps in the domain
IP
The causal maps for partial relations of IPR protection ontology are shown in Fig. 5, which has been constructed through a study of a literature review. In order to represent the relations simply, the annotations used in Fig. 5 are shown in Table 1. The IPR protection methodologies protect the IP works in their lifecycle by affecting some primitives. From the IP protection point of view, the two primary primitives are the security of the IP works and the copyright proof of IP works (shown in Table 2). For example, the access control methodology can increase the security of IP works, while it has no effect on copyright proof of IP works. By contrast, the copyright watermark methodology can increase the copyright proof of IP works, while it has no effect on the security of IP works. With regard the overhead in each stage of the IP lifecycle, there is a different requirement of security and copyright proof. For example, in the Creation stage, the security issue precedes the copyright proof issue, whereas in the Assertion stage, the situation
?IP.Sec ?IP.Pro
The security of IP works The copyright proof of IP works
Table 3 Dataproperty of variables of pirate method. PM ?P.Ri ?P.Cos ?P.Pro
The risk of pirate method The cost of pirate method The successful probability of pirate method
is reversed. In the distribution stage and usage stage, both security and copyright proof are equally important.
Fig. 5. Causal maps for partial relations of IPR protection ontology.
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
p7 ¼ fðUCD; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
Table 4 Evaluation of IPR prevention methodology.
Methodology Robustness Efficiency Compatibility Overhead IP lifecycle stage applied AC E MD
+ + +
+ +
+ + ++
+ +
CW
++
++
++
AW
+
++
++
UCW
++
++
++
DC
+
++
+
+
DS
+
++
+
+
DF
+
+
+
SD
+
+
++
Creation Creation Creation, assertion Assertion, distribution, usage Assertion, distribution, usage Distribution, usage Assertion, distribution, usage Assertion, distribution, usage Assertion, usage Creation, usage
The primitives that affect the successful probability of the pirate method are the cost of piracy and the risk of piracy (shown in Table 3). For example, the high security of IP works in the creation stage will increase the cost of piracy; the robust copyright proof in the usage stage will increase the risk of piracy; both situations will reduce the successful probability of piracy acts: A causal map is a pair: M ¼ ðC; EÞ. C ¼ fAC; CW; ?IP:Sec; ?P:Cos; . . . . . .g. E ¼ fðAC; ?IP:Sec; þÞ; ðAC; ?IP:Pro; 0Þ; ð?P:Co; ?P:Pro; þÞ; . . . . . .g. There are many paths from the objects in IPR Protection Methodologies class to the ‘‘?P.Pro’’ of object in the Pirate method class in Fig. 5. These paths can be written in rules, for example, (a = AC, b = ?P.Pra), p1 ; p2 P a;b : þ
þ
p1 : ðACÞ !ð?IP:SecÞ !ð?P:CosÞ !ð?P:ProÞ p1 ¼ fðAC; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p1
d¼
þ
þ
p2 : ðEÞ !ð?IP:SecÞ !ð?P:CosÞ !ð?P:ProÞ p2 ¼ fðE; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p2
d¼
p3 ¼ fðMD; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p3
d¼
p4 ¼ fðDF; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p4
d¼
p5 ¼ fðCW; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p5
d¼
p6 ¼ fðAW; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p6
d¼
1397
ðc;c0 ;dÞ2p7
d¼
p8 ¼ fðDC; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p8
d¼
p9 ¼ fðDC; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p9
d¼
p10 ¼ fðDS; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p10
d¼
p11 ¼ fðSD; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p11
d¼
Paths set {p1, p2, p9, p11} affect the successful probability of piracy by increasing the cos of pirate methodology, while paths set {p3, p4, p5, p6, p7, p8, p10} affect the successful probability of piracy by increasing the risk of pirate methodology, for these methodologies are able to provide proof of piracy. 5. Case study This case concerns a famous copyright violation case in China (Sina News, 2009). Tomato Garden is a pirate operating system of which the prototype is Windows XP. Before the principal offender HL was sent to prison, Tomato Garden was easily found in computer shopping malls in China. Fig. 6 is used to illustrate the Tomato Garden case from the ontological view. All the entities in Fig. 6 are properties of individual instances of the classes in the ontology in Section 3. (1) As a creator in the business domain, Microsoft created the famous operating system Windows XP on 25 October 2001. It claimed the IP right of Windows XP and could transfer the distribution rights to distributors through rights transfer. Customers purchase the Windows XP operating system from distributors to install on their computers, i.e. customers obtained a usage license through this legal channel. (2) The principal offender, HL, used IP tools and methodologies such as programing software and reverse engineering, cracked Windows XP, and modified the appearance of the Windows XP operating system into Tomato Garden. He then provided free download of Tomato Garden on his website. (3) Certain pirate corporations packaged Tomato Garden and sold it for a very low price. (4) Some potential customers for Windows XP found that Tomato Garden could be obtained free or at very low cost, whereas Windows XP was considerably more expensive than Tomato Garden. These potential customers therefore elected to acquire Tomato Garden through the two pirate channels rather than buy Windows XP from authorized distributors. (5) Tomato Garden led to a significant decline in sales of Windows XP in the Chinese market. As a result, the profits of Microsoft and its distributors were reduced. Meanwhile the tax office encountered loss of revenue. (6) Microsoft appealed to the court; the police carried out investigations and gathered the evidence of piracy. The court made its judgment according to the proof provided by the police and Microsoft.
1398
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 6. Time sequential diagram of tomato garden case.
(7) On 15 August 2008, the principal offender, HL was sentenced to prison for 3 years and 6 months and fined 1million Yuan. (8) In order to prevent this type of piracy recurring, Microsoft has devoted more effort to inventing IPR protection methodologies which enhance the security of its IP products. Meanwhile, the Chinese government has increased its vigilance with regard to pirate sellers in order to protect legal commercial participants.
6. Conclusions A wide variety of industry standards and technologies have emerged in recent years to address the set of IPR protection related issues previously described. Here, we outline and study a representative collection of current IPR protection methodologies by evaluating their effectiveness in addressing the relevant piracy risks. Our choices of current IPR protection methodologies are by no means
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
complete; they merely serve as examples of current industry piracy protection standards and recommendations. Should new technologies and approaches emerge, such approaches will need to be assessed against pirate risks, and technology framework devised to further enhance the classes of IPR protection methodologies. Our framework aims to provide a means to gauge the appropriate methodologies to protect IP works against the class of pirate methodologies. Our evaluation of the individual IPR protection methodologies and their performance in robustness, efficiency, compatibility and overhead are denoted by the following symbols in Table 4: A single plus sign (+) indicates the particular methodology performs well; a double plus sign (++) indicates the particular methodology performs very well; a negative sign () indicates the methodology does not perform as well. Furthermore, the appropriate IP lifecycle stages that each methodology applies to are shown in the table. In this paper, ontologies illustrating the knowledge about IPR protection are proposed. The ontologies comprise three parts. The first part provides a static ontology for the domain language, including primarily classes of participants, classes of IP works, classes of activities, and relations between these classes. In the second part, a dynamic ontology illustrates the IPR protection process. Finally, a causal map is used to demonstrate how classes of IPR protection methodology are causally related with classes of IP piracy methodology. There are three contributions by our study, on the basis of the ontology:
1399
(1) It provides a comprehensive understanding of the domain knowledge about IPR protection. (2) It helps to provide the building blocks for IPR protection models. (3) It facilitates and supports the IPR protection systems’ design and development in this domain. The IPR protection issue in today’s world is obviously much more complex than discussed in this study. The purpose here is to show how ontologies can be built from different aspects of the IPR protection issue and how they can be implemented and used together. A limitation of this study is that the proposed ontologies are only evaluated through the Tomato Garden case. In future, more complex real world cases will be studied in order to refine and complete these ontologies, and an IPR management system will be designed to provide an effective solution to IPR violation. Acknowledgements This research is funded and supported by the National High Technology Research and Development Program of China (863 Program), the Natural Science Foundation of Hubei Province, No. 2008CDA020, and the Independent Innovation Fund of Wuhan University of Technology. Appendix A OWL representation of classes
. . .. . .
1400
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
References Camp, L. J. (2003). First principles of copyright for DRM design. IEEE Internet Computing, 7(3), 59–65. Chaib-draa, B. (2002). Causal maps: Theory, implementation, and practical applications in multiagent environments. IEEE Transactions on Knowledge and Data Engineering, 14(6), 1201–1217. Chandrasekaran, B., Josephson, J. R., & Benjamins, V. R. (1999). What are ontologies, and why do we need them? IEEE Intelligent Systems, 14(1), 20–26. Chen, C. L. (2008). A secure and traceable E-DRM system based on mobile device. Expert Systems with Applications, 35, 878–886. Cohen, J. E. (2003). DRM and privacy. Communications of the ACM, 46, 46–49. Delgado, J., Gallego, I., Llorente, S., & García, R. (2003). IPROnto: An ontology for digital rights management. In The sixteenth annual conference on legal knowledge and information systems, JURIX 2003. Amsterdam: IOS Press. Felten, E. W. (2003). A skeptical view of DRM and fair use. Communications of the ACM, 46, 56–59. García, R., & Gil, R. (2006). An OWL copyright ontology for semantic digital rights management. In Paper presented at the OTM workshops 2006. García, R., & Gil, R. (2010). Content value chains modeling using a copyright ontology. Information Systems, 35(2010), 483–495. Horrocks, I., Patel-Schneider, P. F., Bobley, H., Tabet, S., Grosof, B., & Dean, M. (2004). SWRL: A semantic Web rule language combining OWL and RuleML. In W3C (Ed.). Horrocks, I., Patel-Schneider, P. F., Bechhofer, S., & Tsarkov, D. (2005). OWL rules: A proposal and prototype implementation. Journal of Web Semantics, 3(1), 23–40. Iannella, R. (2002). Open digital rights language (ODRL), Version 1.1, W3C Note, 19 September 2002.. Accessed 12.04.10. Jamkhedkar, P. A., & Heileman, G. L. (2008). Digital rights management architectures. Computers and Electrical Engineering, 35(2009), 376–394. Jurisica, I., Mylopoulos, J., & Yu, E. (2004). Ontologies for knowledge management: An information systems perspective. Knowledge and Information Systems, 6(2004), 380–401. Kaplan, M. A. (1996). IBM cryptolopes, superdistribution and digital rights management. . Accessed 10.06.07. Koenen, R. H., Lacy, J., MacKay, M., & Mitchell, S. (2004). The long march to interoperable digital rights management. Proceedings of the IEEE, 92(6), 883–897.
Kohl, U., Lotspiech, J., & Kaplan, M. A. (1997). Safeguarding digital library contents and users: Protecting documents rather than channels. D-Lib Magazine, 3(9). Ku, W., & Chi, C. H. (2004). Survey on the technological aspects of digital rights management. Proceedings of the Information Security, 3225, 391–403. Lee, B. R. (2007). Mobile broadcast DRM based on user identity card. EURASIP Journal on Wireless Communications and Networking, 7. Lee, J., Hwang, S. O., Jeong, S. W., Yoon, K. S., Park, C. S., & Ryou, J. C. (2005). A DRM framework for distributing digital contents through the Internet. ETRI Journal, 25(6), 423–436. MPEG, (2002). MPEG-21 overview, ISO/IEC JTC1/SC29/WG11/N5231.. Accessed 21.03.10. MPEG, (2003). MPEG-21 rights expression language (REL), ISO/IEC JTC 1/SC 29/WG 11/N5839. . Accessed 21.03. 10. MPEG, (2003). MPEG-21 rights data dictionary (RDD), ISO/IEC JTC 1/SC 29/WG 11/ N5842. . Accessed 21.03.10. Reisman, A., & Heights, S. (2006). A taxonomic view of illegal transfer of technologies: A case study. Journal of Engineering Technology Management, 23(2006), 292–312. Sina News, (2009). The author of Tomato Garden was under arrest. . Accessed 20.08.09. Smith, M. K., Welty, C., & McGuiness, D. L. (2004). OWL Web ontology language guide. . Accessed 05.01.07.
Contents lists available at SciVerse ScienceDirect
Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa
Ontologies for intellectual property rights protection Zhang X.M. a,⇑, Liu Q. a, Wang H.Q. b a b
School of Information Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan, Hubei Province, PR China Department of Information Systems, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
a r t i c l e
i n f o
Keywords: Ontology Intellectual property rights (IPR) Copyright protection Ontology Web Language (OWL)
a b s t r a c t Pirating various forms of intellectual property (IP) causes great economic loss to intellectual property rights (IPR) holders. IPR protection is becoming a key issue in our highly networked world. In order to further deepen our understanding of how to protect IPR and enhance information interchange and knowledge sharing among related entities, ontologies for IPR protection are proposed. This study contains three parts, which are developed to deal with different perspectives in this domain. The first part presents a static ontology, i.e. a hierarchy framework for the domain language, including primarily classes of participants, classes of IP works, classes of activities, and relations between these classes. In the second part, a dynamic ontology is shown to illustrate the IPR protection process. Thirdly, a causal map is used to demonstrate how classes of IPR protection methodologies are causally related with classes of IP piracy methodologies. Finally, the case of Tomato Garden is offered to demonstrate how the proposed ontologies are used in the real world. In respect of the ontology, it is first helpful to gain a comprehensive understanding of domain knowledge of IPR protection; second, IPR protection systems’ design and development in this domain are facilitated and supported by these ontologies; third, the proposed ontologies are united in the Ontology Web Language (OWL) and the OWL rules languages framework, both of which are machine readable. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction The World Intellectual Property Organization (WIPO), defines intellectual property (IP) as creations of the mind: inventions, literary and artistic works, and symbols, names, images, and designs used in commerce. IP is divided into two categories: Industrial property, which includes inventions (patents), trademarks, industrial designs, and geographic indications of source; and Copyright, which includes literary and artistic works such as novels, poems and plays, films, musical works, artistic works such as drawings, paintings, photographs and sculptures, and architectural designs (WIPO, 2004). IP allows people to own their creativity and innovation in the same way that they can own physical property. The owner of IP can control and be rewarded for its use, and this encourages further innovation and creativity to the benefit of everyone. Often, more than one of the above protection types may apply to the same creation. Innovation in information technologies and network communications offers people a great opportunity for the widespread and efficient utilization of IP works through various channels. As well as enjoying the convenience of worldwide information sharing, however, the entire society is faced with the issue of violation of ⇑ Corresponding author. Tel.: +86 2787298927; fax: +86 2787665520. E-mail address: [email protected] (X.M. Zhang). 0957-4174/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2011.08.021
IPR. IPR violation overlaps with issues of commercial domain, legal domain and technical domain – piracy of software, audio, database, books, reverse engineering of marketed hardware as well as theft of sensitive commercial designs by competing corporate entities. The piracy of IP works is a major form of IP violation. The International Intellectual Property Alliance (IIPA) estimated the annual loss of revenue in the US business software industry due to piracy at US$14273 million, and in the record and music industries at US$1486.9 million, for the financial year of 2009, as reported on 18 February 2010 (IIPA, 2010). It is also worth noting that a large portion of Internet bandwidth (approximately 30%) is consumed by users exchanging illegal copies of digital media (mainly video). It is certain that there will always be people with enough motivation to illegally use IP works by circumventing protection mechanisms (Vassiliadis & Fotopoulos, 2007). It is the goal of our paper to propose ontologies that illustrate the domain knowledge about IPR protection. The paper is comprised of three parts. The first part, which is represented using the description logic variant of the Web Ontology Language (OWL DL), provides a static ontology, i.e. a hierarchy framework for the domain language, including primarily classes of participants, classes of IP works, classes of activities, and relations between these classes. It constitutes a specification of the domain-specific concepts of classes, entities, properties, and activities as a set of relationships that exist among these vocabulary terms. In the second part, a dynamic ontology is presented to illustrate
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
the IPR protection process. Thirdly, a causal map is used to demonstrate how classes of IPR protection methodology are causally related with classes of IP pirate methodology, which can be written as rules using the OWL rules language (Horrocks, Patel-Schneider, Bechhofer, & Tsarkov, 2005). The ontology in the paper provides not only a formal description of objects in the domain knowledge and shared terminology, but also a formal basis for reasoning domain knowledge. Thus on the basis of the ontologies, it is first helpful to gain a comprehensive understanding of the domain knowledge of IPR protection; second, IPR protection systems’ design and development in this domain are facilitated and supported by this ontology. The proposed ontologies are then united in the OWL and the OWL rules languages framework, both of which are machine readable, part of which is shown in Appendix A. The rest of the paper is organized as follows: Section 2 illustrates the related techniques, i.e. ontology, OWL DL, and OWL rules language; the details of static and dynamic ontologies are presented in Section 3. The causal map in the domain will be proposed in Section 4. In Section 5, the Tomato Garden case is analyzed using the proposed ontologies. Finally, conclusions are presented in Section 6.
2. Background 2.1. Related works In order to solve the problem of IPR violation, many digital rights management (DRM) systems are proposed in literature. Some significant references include: Camp (2003), illustrating first principles of copyright for DRM design; Torres, Serrao, Dias, and Delgado (2008), offering an analysis of the various methods for implementing interoperable digital rights management platforms; Jamkhedkar and Heileman (2008), analyzing the problems with current DRM environments and proposing an open layered framework for the development of DRM systems; Thomas, Emmanuel, Subramanyam, and Kankanhalli (2009) proposing a joint digital watermarking scheme using the Chinese remainder theorem for the multiparty multilevel DRM architecture; and Lee et al. (2005), for designing a contents distribution framework that supports transparent distribution of digital contents on the Internet as well as the copyright protection of participants in the contents distribution value chain. A useful analysis of DRM business models, standards, and core technologies can be found in Koenen, Lacy, MacKay, and Mitchell (2004), Ku and Chi (2004), Cohen (2003), and Felten (2003). The increasing use of mobile devices has also initiated research efforts for mobile DRMs (MDRMs); technological challenges in this area differ from classic DRM and include mobile device limitations, bandwidth, usability, among others (Chen, 2008; Lee, 2007). Many companies, organizations and administration-funded projects provide solutions for the implementation of DRM systems, which has given rise to a boom in commercial DRM systems, such as the Adobe e-book for pdf documents, the IBM cryptolope (Kaplan, 1996), the ambitious DigiBox technology by InterTRust (Kohl, Lotspiech, & Kaplan, 1997), Microsoft’s Windows Media Player for audio/video, and Digimarc’s family of products for video/audio and still images, to name a few. Nevertheless, although these solutions have several aspects in common, they are incompatible in terms of architecture and system components. Moreover, efforts that facilitate copyright management in closed domains experience great difficulty when they are forced to interoperate in an open domain like the World Wide Web. In order to facilitate interoperation and automation, DRM systems can be enriched with domain formalizations. There are many other initiatives to build an IPR framework for the Internet-wide management of IP works; for instance, MPEG21 (MPEG, 2002) or, in the W3C (World Wide Web Consortium)
1389
initiatives framework, the ODRL (Open Digital Rights Language) proposal (Iannella, 2002). Most of these initiatives have one thing in common: they work at the syntactic level. Their approach is to formalize XML DTDs (Document Type Definition) and Schemas that define a rights expression language (REL), such as the MPEG-21 REL (MPEG, 2003). In some cases, the semantics of these languages, i.e. the meaning of the expressions, are also provided but formalized separately as rights data dictionaries (RDD); for instance, MPEG21 RDD (MPEG, 2003). Rights dictionaries define terms in natural language, solely for human consumption; for this reason it is not easy to automatically process those terms. However, the syntactic approach does not scale well in really wide and open domains like the Internet. The automatic processing of a huge amount of metadata coming from many different sources requires machine understandable semantics. The syntax is inadequate when unforeseen expressions are encountered, which is where semantics can assist with their interpretation to achieve interoperation. Delgado, Gallego, Llorente, and García (2003), proposed an ontology for DRM which concentrates only on some aspects of IPR protection, in particular, the concepts of IPR Agreement, including Contract and License, and some specific rights, such as exploitation or moral rights. A copyright ontology is implemented in García and Gil (2006), using OWL DL. This approach facilitates the implementation of efficient usages against license checking, which is reduced to description logics classification. In order to support the whole copyrighted content value chain across enterprise or business niches boundaries, García and Gil (2010) provides a framework that accommodates copyright law and a rich creation model in order to cope with all the creation life cycle stages. The dynamic aspects of value chains are modeled using a hybrid approach that combines ontology-based and rule-based mechanisms. These ontologies were constructed from a rights management point of view; in particular, the concepts of Contract and License. Nevertheless, there is a need for a more comprehensive ontology to illustrate the IPR protection issue.
2.2. Ontology As a branch of philosophy, ontology is the study of the kinds of things that exist. It is often said that ontologies ‘carve the world at its joints’ (an expression commonly attributed to Plato). In the artificial intelligence (AI) community, the term ontology is used to refer to a set of representation vocabulary, and, more precisely, it is the conceptualizations in a domain that the terms in the vocabulary are intended to capture (Chandrasekaran, Josephson, & Benjamins, 1999). According to Sowa, knowledge representation can be defined as the application of logic and ontology to the task of constructing computable models of a domain (Sowa, 2000). Concerning the nature and relations of being, ontologies can be used as critical components in knowledge management and have huge potential to improve information organization, management, and understanding. In order to further deepen our understanding of IPR protection issues and enhance information and knowledge sharing among related entities, ontologies for IPR protection issues are proposed in this study. Jurisica, Mylopoulos, and Yu (2004), have classified ontologies into four broad categories: static, dynamic, intentional, and social. Two of these ontologies will be considered here: those that deal with static and dynamic aspects of IPR protection issues. Static ontologies represent the static aspect and define the basic concepts of IPR protection issues, while dynamic ontologies deal with knowledge regarding the IPR protection process. This paper will also discuss the implementation level at which OWL and OWL rules languages are used to represent these ontologies in a machine-readable form.
1390
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
2.3. OWL and OWL rules languages OWL is a good representation for ontology and was developed by the world standards body, W3C (Smith, Welty, & McGuiness, 2004). It is designed for use by applications that need to process the content of information rather than simply presenting information. It allows more vocabulary terms to be used for describing properties and classes. The main advantages of OWL are efficient reasoning support, sufficient expressive power, and convenient expression. OWL offers three sublanguages: OWL Lite, OWL DL, and OWL Full. OWL DL supports those users who want maximum expressiveness while retaining computational completeness and decidability. In our study, OWL DL is suitable for representing the various types of classes and relationships in the ontology. Although the OWL Web Ontology Language has considerable expressive power, it also has expressive limitations, particularly with respect to what can be said about properties. In order to overcome these limitations, Horrocks et al. (2004) proposed the Semantic Web Rules Languages (SWRL). SWRL extends OWL with a simple form of Horn-style rules (Horrocks et al., 2004). The proposed rules are of an implication form between an antecedent (body) and consequent (head). Both the antecedent and consequent of a rule consist of zero or more atoms. The informal meaning of a rule can be read as: whenever the conditions specified in the antecedent hold, then the conditions specified in the consequent will also hold. Atoms in rules can be of the form C(x), P(x, y), Q(x, z), same as (x, y), or different (x, y), where C is an OWL DL description, P is an OWL DL individual-valued Property, Q is an OWL DL data valued Property, x and y are either variables or OWL individuals, and z is either a variable or an OWL data value. 3. Ontologies for IPR protection In this section, ontologies for IPR protection will be given, and will include static and dynamic ontologies. 3.1. Static ontology As noted above, the static ontology represents the static aspect of the IPR protection issue and defines the basic concepts for that issue, e.g. IP works, information technology, corporate and so forth. As analyzed in Section 1, the IPR domain includes various kinds of entities, and there are complex relationships among them. Fig. 1(a) shows the top-level ontology of IPR protection, through which general knowledge is given. Three classes are shown in the top level of the static ontology: the IP works class, the participants class, and the activities class. The IP works class comprises various forms of IP that are the most likely to be violated, namely, the IP distribution channel and IP lifecycle. The participants class is used to represent the roles involved in IPR protection issues. According to the interest gained from IP Works, the roles are classified into three subclasses: pirate, anti-pirate and neutrality. Each of them has individual, corporate and other types. The activities class describes the actions taken by the participants. It mainly involves an IPR protection methodology subclass, which provides the essential approaches supporting IPR protection, and the piracy methodology subclass, which presents the most commonly used piracy approaches. These three classes have complex relationships; in addition, their subclasses also have intricate relationships with each other. The details of these relationships at the class level will be discussed in subsequent sections. Fig. 1(b) shows the subclass of the form of IP works in detail. The IP works class is made up of various forms of IP that are most likely to be violated. The act of piracy is most likely to be committed against the three main subclasses of IP works: multimedia,
software and design. The most common types of multimedia include text, image, audio, video, and similar forms. Operating systems and applications are the two main software types. Designs such as blueprints, circuit design, database, medicinal compounds, cosmetics, etc., make up another form that is highly likely to be violated. Fig. 1(c) illustrates the primary distribution channels of IP works. Traditionally, the physical distribution channels are CD, DVD, Book, or some other fixation medium subject to copyright. With the extensive use of networks, the online shop has become a very popular distribution channel for IP works. Another distribution channel through networks is that of super distribution by P2P file sharing, email, instant messaging, and so on. Fig. 1(d) describes the lifecycle of IP works. There are four main stages in the lifecycle: the creation stage, assertion stage, distribution stage and usage stage. IPR protection methodology class is shown in Fig. 1(e). Its subclasses are all possible techniques to protect IPR or prevent piracy. The most commonly used techniques are encryption algorithms, such as DES, AES, RSA, Elliptic Curve and so forth; metadata; digital certificates, such as X509; digital signature, such as RSA signature or hash signature; digital watermarking – a promising solution to copyright problems – such as copyright watermark, authentication watermark, and usage control watermark. Finally, the piracy methodology class is presented in Fig. 1(f). The most commonly used pirate methods are copying, downloading, reuse, dishonest claims, superdistribution, rerecord, and crack.
3.2. Relationships between participants and activities In Fig. 2, the Participants class is classified into three types: Pirate, Anti-Pirate and Neutrality. These participants represent the commercial entities in the real world. Piracy occurs in one of two ways: naive or malicious (Reisman & Heights, 2006). The way in which it occurs is determined by the motivations of commercial entities. The pirate class is composed of corporate entities, institutions, individuals A company may intend to pirate the IP of its opponent in order to gain greater profit and outmaneuver its commercial competitors. An individual may crack software and share it with friends. An institution may install unauthorized software on its computers. Whether the piracy takes place with naive or malicious motivations, it results in loss of profit for the IPR holders, who mainly fall into the anti-pirate category. The anti-pirate category consists of government, institutions, corporate entities, and individuals. All the members of this group try to prevent or reduce piracy and, if necessary, to punish the pirate. Neutrality exists as a third category, which is most likely to be made up of potential customers of certain IP works. Once conditions change, a customer may turn to either of the other two categories. For example, Alice wants to buy a CD from an online shop. If she finds the CD can be obtained free on a website, then she is most likely to change her mind and download the CD from the website. The left side of this figure shows the classes of Participants. The right side shows the classes of activities. The dashed open arrows link two different classes, and denote their Objectproperties. For example, if, at the class level, the arrow labeled ‘‘adopted by’’ links a Participants class and an Activities class, it means that the objects in the Activities class are adopted by the objects in the Participants class. Meanwhile, the arrow labeled ‘‘Invent’’ means that the objects in the Participants class invent the objects in the Activities class. For example, at the instance level, a copyright watermark is embedded into original data to identify its ownership by the author, John. The distributor, ADS Ltd, can adopt an X.509 Certificate to protect the content in the distribution stage. The pirate company, HH, produces a pirate CD by using Combo CD writer.
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
1391
Fig. 1. Class of static ontology. (a) Class of static ontology. (b) IP works class. (c) IP distribution channel class. (d) IP lifecycle class. (e) IPR protection methodology class. (f) Piracy methodology class.
1392
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 2. Relationships between participants and activities.
The famous encryption algorithm, RSA was invented by Ron Rivest, AdiShamir, and Leonard Adleman. 3.3. Relationships between Participants and IP Works As shown in Fig. 3, the left side in the figure is the classes of IP works. IP works in different forms are distributed through different IP distribution channels. Each IP work will go through an IP lifecycle, including a creation stage, assertion stage, distribution stage and usage stage. The right side of this figure is the classes of Participants. As shown at the class level, the arrow labeled ‘‘take activities around’’ links the Participants class and IP works class, which means that the objects in the Participants class take activities around the objects in the IP works class. Meanwhile, the arrow labeled ‘‘have effect on’’ means that the objects in the IP works class have an effect on the objects in the Participants class. For example, at the instance level, a potential customer may pirate anti-virus software distributed by P2P file sharing. A pirate company intends to pirate the circuit design through the manufacturer. A creator is active in the creation stage of the IP lifecycle and corporate distributor is active in the distribution stage of the IP lifecycle. 3.4. Dynamic ontology of IPR protection Dynamic ontologies describe changing aspects of the world. Typical primitive concepts include state, state transition and process. The dynamic view of IPR protection allows the construction of IPR protection models for use in electronic commerce. This includes the possibility of defining events in the IP works lifecycle. In this section, the dynamic ontology deals with knowledge of the IPR protection process and a UML state diagram, shown in Fig. 4, is adopted to depict it.
In order to provide a detailed description of the IPR protection process, Fig. 4 has several parts. Fig. 4(a) shows that the general IP protection process consists of five parts: the creation process, assertion process, usage process and monitoring process. Fig. 4(b1) shows the IPR protection process in the creation stage. At the beginning of the process, a creator creates content (creation stage). If the creator wants to protect his creation from piracy, he can apply for IPR protection methodology. In this stage, the creation data is saved on a computer. As a result, different IPR protection methodologies are selected to protect the computer and the creation data. Here for instance, access control is used to protect the computer from unauthorized access, and encryption is used to protect the creation data. After this deployment, the creation data is in a protected state. Fig. 4(b2) shows the pirate process in the creation stage. A pirate who wants to steal the creation data in this stage will apply for appropriate pirate methodologies. In the situation that creation data saved on a computer, the pirate firstly illegal accesses the computer. If the computer is protected by an access control policy, it will check whether the pirate access rights. Once the rights check confirms this is an illegal access, the pirate attempt is prevented. In the event that the rights check confirms an authorized access, the pirate will enter the operating system and apply other pirate methodologies to obtain the creation data. If the creation data is encrypted by the creator, the pirate will try to decrypt it. Only if the decryption is successful can the pirate obtain the creation data by copying or downloading. Fig. 4(c1) illustrates the IPR protection process in the assertion stage. After completion of the creation, the creator should assert content when it is first created (or is reused and extended with appropriate rights). Firstly, the creator has to submit the content and metadata to the authority for request. The authority applies for appropriate protection methodologies, i.e. digital fingerprinting
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
1393
Fig. 3. Relationships between participants and IP works.
to identify the content, and to ensure that the same content is not already registered. If the same content exists, the assertion fails. Otherwise, the authority confirms this is novel content and the creation is validated. The next step is rights assertion. The creator makes the rights request to the authority and the latter applies for IPR protection methodologies, i.e. the generation of a digital watermark/certificate, which is distributed to the creator to identify him or her as the rights holder. Fig. 4(c2) describes the pirate process in the assertion stage. The most likely scenario is that the pirate has obtained a copy of the creation data at an early stage and then dishonestly claims ownership of this creation to the authority. Through checking digital fingerprinting/watermarking, the authority identifies whether the content is a novel creation, in Fig. 4(c1). If the real creator has successfully asserted rights, the pirate will be prevented from proceeding. The IPR protection process in the distribution stage is shown in Fig. 4(d1). The creator can make a request to the authority to transfer the IPR. The authority applies for IPR protection methodologies, generating digital certificate/watermarking and distributing it to the rights holder and distributor. Meanwhile, the creator creates a business model for the IP works, which is codified as rules. The content and codified rules are packaged with encryption/certificate/watermarking and prepared for distribution. The distributor is responsible for managing and enabling the trade of the content. In order to access the content, the customer has to complete payment through the authority. On receiving payment from the customer, the authority confirms this to the distributor, who generates the digital certificate/watermarking which carries the usage rights and distributes it to the customer. The pirate process in the distribution stage is shown in Fig. 4(d2). At the start of the process, the pirate makes an application to the distributor to view the content and obtains the watermarked content/temporal certificate. The pirate then applies pirate methodologies to complete the theft by, for instance, rerecording the content or keeping an illegal copy of the content. Final-
ly, the pirate obtains watermarked content, which exposes him/her to high risk. Fig. 4(e1) illustrates the IPR protection process in the usage stage. In the initial stage, the customer obtains encrypted/watermarked content and the end-user player in the distribution stage. Before the content is used by the customer, the IPR protection methodologies are carried out to check whether the customer holds the legal usage rights. If the check fails, usage is forbidden. Fig. 4(e2) illustrates the pirate process in the usage stage. This type of piracy is most likely to impact a customer who has purchased authorized content. The pirate applies pirate methodologies to copy/download/crack the customer’s authorized content and obtains a copy of the content. The pirate then reuses the content, packages and sells the content, or shares the content with others through superdistribution, or similar means. The IPR protection process in the monitoring stage is shown in Fig. 4(f). After selling the content to the customer, the distributor will monitor usage through IPR protection methodologies. The most commonly used methodology is agent technique. The crawling agent searches suspicious content in the networks, checks whether there is a digital watermark/fingerprint/certificate in the content, makes comparison with the record database and judges whether the content has been pirated or is unauthorized. A distributor who has uncovered illegal usage has the right to bring legal action against the pirate.
4. Causal maps in the domain 4.1. Causal maps A causal map offers to model interrelationships among a variety of concepts; it can be employed to cope with complicated problems for which analytical techniques are inadequate. It is one of five generic types of cognitive map, and is still the most popular mapping method (Chaib-draa, 2002). When a causal map is
1394
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 4. Dynamic ontology for IPR protection. (a) IPR protection process for IP works. (b1) IPR protection process in creation stage. (b2) Pirate process in creation stage. (c1) IPR protection process in assertion stage. (c2) Pirate process in assertion stage. (d1) IPR protection process in distribution stage. (d2) Pirate process in distribution stage. (e1) IPR protection process in usage stage. (e2) Pirate process in usage stage. (f) IPR protection process in monitoring stage.
pictured in graph form, it is relatively easy to see how concepts and causal relationships are related to each other and to see the overall causal relationships of one concept with another. A causal map M is a pair, M = (C, E), where C is a set of nodes, or concepts, and E is a set of signed edges (Wellman, 1994). E = {(c, c0 , d): c, c0 e C, there is an edge of sign d from c to c0 }. The possible signs are usually ‘‘+’’, ‘‘’’, ‘‘0’’ and ‘‘?’’. They respectively denote ‘‘c has increasingly effect on c0 ’’, ‘‘c has decreasingly
effect on c0 ’’, ‘‘c has no effect on c0 ’’, and ‘‘c has ambiguous effect on c0 ’’. Usually, there are two types of concept events in causal maps: existential and directional. The directional event is used in this paper. For c 2 C, c is a directional event. Positive change, denoted by c(+), represents it is known that c is rising.
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 4 (continued)
1395
1396
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Zero change, denoted by c(0), represents it is known that c is unchanged. Negative change, denoted by c(), represents it is known that c is decreasing. d
þ
(c, c , d)((c, c , d) e E) can be written as rule: c ! c . c ! c means that if c is ‘‘+’’(‘‘’’), then c0 will hold as ‘‘+’’(‘‘’’). c ! c0 means that 0 if c is ‘‘+’’(‘‘’’), then c0 will hold as ‘‘’’(‘‘+’’). And c ! c0 means that c ? has no effect on c0 . c ! c0 means that c has ambiguous effect on c0 . A causal map graph is a set of classes connected by arrows, with the direction of the arrow indicating the direction of influence or causality. The general definition of derived relationships between in an acyclic causal map is simply the sum of the signs for all paths between them. Let Pa,b be the set of paths in M from a to b, then the sign of derived causal relation from a to b is: p2Pa;b ððc;c0 ;dÞ2p dÞ. 0
0
0
0
Table 1 The annotations used in Fig. 5. IPR protection method AC E MD CW AW UCW DC DS DF SD
Access control Encryption Metadata Copyright watermark Authentication watermark Usage control watermark Digital certificate Digital signature Digital fingerprinting Specific device
Table 2 Dataproperty of variables of IPR protection method.
4.2. Causal maps in the domain
IP
The causal maps for partial relations of IPR protection ontology are shown in Fig. 5, which has been constructed through a study of a literature review. In order to represent the relations simply, the annotations used in Fig. 5 are shown in Table 1. The IPR protection methodologies protect the IP works in their lifecycle by affecting some primitives. From the IP protection point of view, the two primary primitives are the security of the IP works and the copyright proof of IP works (shown in Table 2). For example, the access control methodology can increase the security of IP works, while it has no effect on copyright proof of IP works. By contrast, the copyright watermark methodology can increase the copyright proof of IP works, while it has no effect on the security of IP works. With regard the overhead in each stage of the IP lifecycle, there is a different requirement of security and copyright proof. For example, in the Creation stage, the security issue precedes the copyright proof issue, whereas in the Assertion stage, the situation
?IP.Sec ?IP.Pro
The security of IP works The copyright proof of IP works
Table 3 Dataproperty of variables of pirate method. PM ?P.Ri ?P.Cos ?P.Pro
The risk of pirate method The cost of pirate method The successful probability of pirate method
is reversed. In the distribution stage and usage stage, both security and copyright proof are equally important.
Fig. 5. Causal maps for partial relations of IPR protection ontology.
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
p7 ¼ fðUCD; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
Table 4 Evaluation of IPR prevention methodology.
Methodology Robustness Efficiency Compatibility Overhead IP lifecycle stage applied AC E MD
+ + +
+ +
+ + ++
+ +
CW
++
++
++
AW
+
++
++
UCW
++
++
++
DC
+
++
+
+
DS
+
++
+
+
DF
+
+
+
SD
+
+
++
Creation Creation Creation, assertion Assertion, distribution, usage Assertion, distribution, usage Distribution, usage Assertion, distribution, usage Assertion, distribution, usage Assertion, usage Creation, usage
The primitives that affect the successful probability of the pirate method are the cost of piracy and the risk of piracy (shown in Table 3). For example, the high security of IP works in the creation stage will increase the cost of piracy; the robust copyright proof in the usage stage will increase the risk of piracy; both situations will reduce the successful probability of piracy acts: A causal map is a pair: M ¼ ðC; EÞ. C ¼ fAC; CW; ?IP:Sec; ?P:Cos; . . . . . .g. E ¼ fðAC; ?IP:Sec; þÞ; ðAC; ?IP:Pro; 0Þ; ð?P:Co; ?P:Pro; þÞ; . . . . . .g. There are many paths from the objects in IPR Protection Methodologies class to the ‘‘?P.Pro’’ of object in the Pirate method class in Fig. 5. These paths can be written in rules, for example, (a = AC, b = ?P.Pra), p1 ; p2 P a;b : þ
þ
p1 : ðACÞ !ð?IP:SecÞ !ð?P:CosÞ !ð?P:ProÞ p1 ¼ fðAC; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p1
d¼
þ
þ
p2 : ðEÞ !ð?IP:SecÞ !ð?P:CosÞ !ð?P:ProÞ p2 ¼ fðE; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p2
d¼
p3 ¼ fðMD; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p3
d¼
p4 ¼ fðDF; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p4
d¼
p5 ¼ fðCW; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p5
d¼
p6 ¼ fðAW; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p6
d¼
1397
ðc;c0 ;dÞ2p7
d¼
p8 ¼ fðDC; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p8
d¼
p9 ¼ fðDC; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p9
d¼
p10 ¼ fðDS; ?IP:Pro; þÞ; ð?IP:Pro; ?P:Ri; þÞ; ð?P:Ri; ?P:Pro; Þg
ðc;c0 ;dÞ2p10
d¼
p11 ¼ fðSD; ?IP:Sec; þÞ; ð?IP:Sec; ?P:Cos; þÞ; ð?P:Cos; ?P:Pro; Þg
ðc;c0 ;dÞ2p11
d¼
Paths set {p1, p2, p9, p11} affect the successful probability of piracy by increasing the cos of pirate methodology, while paths set {p3, p4, p5, p6, p7, p8, p10} affect the successful probability of piracy by increasing the risk of pirate methodology, for these methodologies are able to provide proof of piracy. 5. Case study This case concerns a famous copyright violation case in China (Sina News, 2009). Tomato Garden is a pirate operating system of which the prototype is Windows XP. Before the principal offender HL was sent to prison, Tomato Garden was easily found in computer shopping malls in China. Fig. 6 is used to illustrate the Tomato Garden case from the ontological view. All the entities in Fig. 6 are properties of individual instances of the classes in the ontology in Section 3. (1) As a creator in the business domain, Microsoft created the famous operating system Windows XP on 25 October 2001. It claimed the IP right of Windows XP and could transfer the distribution rights to distributors through rights transfer. Customers purchase the Windows XP operating system from distributors to install on their computers, i.e. customers obtained a usage license through this legal channel. (2) The principal offender, HL, used IP tools and methodologies such as programing software and reverse engineering, cracked Windows XP, and modified the appearance of the Windows XP operating system into Tomato Garden. He then provided free download of Tomato Garden on his website. (3) Certain pirate corporations packaged Tomato Garden and sold it for a very low price. (4) Some potential customers for Windows XP found that Tomato Garden could be obtained free or at very low cost, whereas Windows XP was considerably more expensive than Tomato Garden. These potential customers therefore elected to acquire Tomato Garden through the two pirate channels rather than buy Windows XP from authorized distributors. (5) Tomato Garden led to a significant decline in sales of Windows XP in the Chinese market. As a result, the profits of Microsoft and its distributors were reduced. Meanwhile the tax office encountered loss of revenue. (6) Microsoft appealed to the court; the police carried out investigations and gathered the evidence of piracy. The court made its judgment according to the proof provided by the police and Microsoft.
1398
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
Fig. 6. Time sequential diagram of tomato garden case.
(7) On 15 August 2008, the principal offender, HL was sentenced to prison for 3 years and 6 months and fined 1million Yuan. (8) In order to prevent this type of piracy recurring, Microsoft has devoted more effort to inventing IPR protection methodologies which enhance the security of its IP products. Meanwhile, the Chinese government has increased its vigilance with regard to pirate sellers in order to protect legal commercial participants.
6. Conclusions A wide variety of industry standards and technologies have emerged in recent years to address the set of IPR protection related issues previously described. Here, we outline and study a representative collection of current IPR protection methodologies by evaluating their effectiveness in addressing the relevant piracy risks. Our choices of current IPR protection methodologies are by no means
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
complete; they merely serve as examples of current industry piracy protection standards and recommendations. Should new technologies and approaches emerge, such approaches will need to be assessed against pirate risks, and technology framework devised to further enhance the classes of IPR protection methodologies. Our framework aims to provide a means to gauge the appropriate methodologies to protect IP works against the class of pirate methodologies. Our evaluation of the individual IPR protection methodologies and their performance in robustness, efficiency, compatibility and overhead are denoted by the following symbols in Table 4: A single plus sign (+) indicates the particular methodology performs well; a double plus sign (++) indicates the particular methodology performs very well; a negative sign () indicates the methodology does not perform as well. Furthermore, the appropriate IP lifecycle stages that each methodology applies to are shown in the table. In this paper, ontologies illustrating the knowledge about IPR protection are proposed. The ontologies comprise three parts. The first part provides a static ontology for the domain language, including primarily classes of participants, classes of IP works, classes of activities, and relations between these classes. In the second part, a dynamic ontology illustrates the IPR protection process. Finally, a causal map is used to demonstrate how classes of IPR protection methodology are causally related with classes of IP piracy methodology. There are three contributions by our study, on the basis of the ontology:
1399
(1) It provides a comprehensive understanding of the domain knowledge about IPR protection. (2) It helps to provide the building blocks for IPR protection models. (3) It facilitates and supports the IPR protection systems’ design and development in this domain. The IPR protection issue in today’s world is obviously much more complex than discussed in this study. The purpose here is to show how ontologies can be built from different aspects of the IPR protection issue and how they can be implemented and used together. A limitation of this study is that the proposed ontologies are only evaluated through the Tomato Garden case. In future, more complex real world cases will be studied in order to refine and complete these ontologies, and an IPR management system will be designed to provide an effective solution to IPR violation. Acknowledgements This research is funded and supported by the National High Technology Research and Development Program of China (863 Program), the Natural Science Foundation of Hubei Province, No. 2008CDA020, and the Independent Innovation Fund of Wuhan University of Technology. Appendix A OWL representation of classes
1400
X.M. Zhang et al. / Expert Systems with Applications 39 (2012) 1388–1400
References Camp, L. J. (2003). First principles of copyright for DRM design. IEEE Internet Computing, 7(3), 59–65. Chaib-draa, B. (2002). Causal maps: Theory, implementation, and practical applications in multiagent environments. IEEE Transactions on Knowledge and Data Engineering, 14(6), 1201–1217. Chandrasekaran, B., Josephson, J. R., & Benjamins, V. R. (1999). What are ontologies, and why do we need them? IEEE Intelligent Systems, 14(1), 20–26. Chen, C. L. (2008). A secure and traceable E-DRM system based on mobile device. Expert Systems with Applications, 35, 878–886. Cohen, J. E. (2003). DRM and privacy. Communications of the ACM, 46, 46–49. Delgado, J., Gallego, I., Llorente, S., & García, R. (2003). IPROnto: An ontology for digital rights management. In The sixteenth annual conference on legal knowledge and information systems, JURIX 2003. Amsterdam: IOS Press. Felten, E. W. (2003). A skeptical view of DRM and fair use. Communications of the ACM, 46, 56–59. García, R., & Gil, R. (2006). An OWL copyright ontology for semantic digital rights management. In Paper presented at the OTM workshops 2006. García, R., & Gil, R. (2010). Content value chains modeling using a copyright ontology. Information Systems, 35(2010), 483–495. Horrocks, I., Patel-Schneider, P. F., Bobley, H., Tabet, S., Grosof, B., & Dean, M. (2004). SWRL: A semantic Web rule language combining OWL and RuleML. In W3C (Ed.). Horrocks, I., Patel-Schneider, P. F., Bechhofer, S., & Tsarkov, D. (2005). OWL rules: A proposal and prototype implementation. Journal of Web Semantics, 3(1), 23–40. Iannella, R. (2002). Open digital rights language (ODRL), Version 1.1, W3C Note, 19 September 2002.
Kohl, U., Lotspiech, J., & Kaplan, M. A. (1997). Safeguarding digital library contents and users: Protecting documents rather than channels. D-Lib Magazine, 3(9). Ku, W., & Chi, C. H. (2004). Survey on the technological aspects of digital rights management. Proceedings of the Information Security, 3225, 391–403. Lee, B. R. (2007). Mobile broadcast DRM based on user identity card. EURASIP Journal on Wireless Communications and Networking, 7. Lee, J., Hwang, S. O., Jeong, S. W., Yoon, K. S., Park, C. S., & Ryou, J. C. (2005). A DRM framework for distributing digital contents through the Internet. ETRI Journal, 25(6), 423–436. MPEG, (2002). MPEG-21 overview, ISO/IEC JTC1/SC29/WG11/N5231.