Semantic Web Publishing Vocabulary (SWP) User Manual
Contents
1. swp digest Method The value is the digest method by which the digest value specified for the graph subject was constructed swp hasKey The value is some kind of public key which belongs to the authority The key is represented by an XML literal containing a XML Signature keyInfo element swp certificate The value is the base64 encoding of a binary ASN 1 DER X 509 certificate containing the public key of the authority Table 1 2 Signature related terms of the SWP vocabulary afterwards Blank node identifier independence is achieved by adding triples that capture the original blank node identifiers to the graph be fore calculating the set hash These triples are used in the signature verification process to temporary relabel blank nodes with their original identifiers Sayers and Karp do not specify a single hash and aggrega tion function for computing the set hash but discuss the advantages and disadvantages of different options For instance SHA1 or MD5 could be used for hashing triples and XOR multiplication or addition could be used for aggregating the resulting hash values into a set hash Both methods can be used to calculate the digest of an RDF graph that is part of a named graph But as a named graph consists of an RDF graph and a graph name the graph name has to be reflected in the digest as well One option to achieve this is to separately digest the graph name and the RDF graph and to combin2. CHAPTER 1 THE SWP VOCABULARY 3 Term Description swp Authority Class of all authorities Information providers as well as information syndicators may act as authorities swp Warrant Class of all warrants A warrant records an authorizing relationship between a graph and an authority swp assertedBy The subject graph is asserted by the authority specified for the object warrant The triples of the graph are taken to be claims made by that authority swp quotedBy The triples of the graph are quoted by the authority and are thus not taken to be claims made by that authority swp authority Defines the authority of a warrant swp validFrom Defines the start of the validity period of a warrant swp validUntil Defines the end of the validity period of a warrant swp sourceURL URL for retrieving a representation of the graph This URL may be used if the graph cannot be retrieved by dereferencing the graph name URI reference Table 1 1 SWP terms for authorizing graphs Table gives an overview about the SWP terms for authorizing graphs The swp authority property relates warrants to authorities The swp assertedBy and swp quotedBy properties capture the propositional atti tude of the relationship between a graph and a warrant These take a named graph as a subject and a swp Warrant as object swp authority takes a warrant as a subject and a swp Authority as an object Each warrant mus
3. ex WarrantedGraphi swp digestMethod swp JjcC14N shal xor 14 ex WarrantedGraphi swp digest 15 qZk NkcGgWq6PiVxeF xsd base64Binary 16 ex WarrantedGraph2 swp assertedBy ex WarrantGraph 17 ex WarrantedGraph2 swp digestMethod swp JjcC14N shal xor 18 ex WarrantedGraph2 swp digest 19 kpRyejYS4uxwT9174F xsd base64Binary 20 ex WarrantGraph swp assertedBy ex WarrantGraph 21 ex WarrantGraph swp authority lt mailto reynolds ft com gt 22 ex WarrantGraph swp signatureMethod swp JjcC14N md5 xor dsa 23 ex WarrantGraph swp signature 24 16GB VsWq5f JKzQcBB4 xsd base64Binary 25 lt mailto reynolds ft com gt swp certificate 26 iVxeFDJO xsd base64Binary 27 Figure 1 4 Signing multiple graphs with a single signature a graph implies a claim by the authority that the content of the graph is true In contrast the Dublin Core terms focus on the role of a person or institution in the process of creating an information resource Thus Dublin Core elements do not imply anything about the truth of created information A second difference lies in the way both vocabularies are used within RDF The Dublin Core working draft Expressing Dublin Core Metadata using the Resource Description Framework RDF specifies that Dublin Core elements are used as predicates of RDF triples describing a resource for instance lt resource gt dc creator Name The SWP vocabulary captures a relationship betwe
4. how the Named Graphs data model and the Semantic Web Publishing vocabulary could be used by a financial information portal to represent financial news analyst reports and newsgroup postings together with quality related meta information such as provenance informa tion ratings and background information about the information providers The developed domain model will be used as a running example for explaining the capabilities of the WIQA Information Quality Assessment Framework in the next part of this thesis 2 1 Domain Model This section describes a domain model for representing financial in formation together with quality related meta information as a set of named graphs The domain model reuses existing terms from the Dublin Core ISO03 FOAF BM06 RDF Schema and ISO 3166 1 vocabularies Domain specific terms are defined in the http www fu berlin suhl bizer 2006 FinVoc namespace which is abbrevi http online wsj com retrieved 09 25 2006 http quote bloomberg com retrieved 09 25 2006 http Finance yahoo com retrieved 09 25 2006 htp Finance google com retrieved 09 25 2006 12 CHAPTER 2 USE CASE fin Security fin Share rdfs label rdfs Literal de description rdfs Literal fin isin rdfs Literal fin emitter fin Company fin news rdfs Literal fin positiveAnalystReport rdfs Literal fin neutralAnalystReport rdfs Literal fin negativeAnal
5. http news ft com gt fin country iso UK lt mailto mark scott com gt rdf type foaf Person foaf name Mark Scott xsd string fin country iso UK fd GraphFromAggregator fd GraphFromIntel swp assertedBy fd IntelWarrant fd IntelWarrant swp authority lt urn x DUNS 047897855 gt dc date 2005 10 21 xsd date fd GraphfromSiemens swp assertedBy fd SiemensWarrant fd SiemensWarrant swp authority lt urn x DUNS 316067164 gt dc date 2005 08 21 xsd date fd GraphFromPeterSmith swp assertedBy fd PsWarrant fd PsWarrant swp authority lt mailto peterSmith deutsche bank de gt dc date 2005 11 20T12 40 44 xsd dateTime fd GraphFromJohnReynolds swp assertedBy fd JrWarrant fd JrWarrant swp authority lt mailto reynolds ft com gt dc date 2005 11 20T09 44 32 xsd dateTime fd GraphFromMarkScott swp assertedBy fd MsWarrant fd MsWarrant swp authority lt mailto mark scott com gt dc date 2005 11 20T17 22 10 xsd dateTime fd fd BackgroundInformation swp assertedBy fd GraphFromAggregator fd GraphFromAggregator swp assertedBy fd GraphFromAggregator swp authority lt urn x DUNS 377667161 gt dc date 2005 11 21T18 20 00 xsd dateTime Figure 2 3 Example Graph Set Part 2 Bibliography BG06 Dan Brickley and Ramanathan Guha RDF Vocabulary Description Language 1 0 RDF Schema W3C Recommendation http www w3 org TR rdf schema 2004 retrieved 09 25 2006 BMO06 Dan Brickley
6. 06 First the digests of all war ranted graphs are calculated These digests are added to the warrant graph as lt NameOfWarrantedGraph gt swp digest DigestValue triples The method that is used to compute a digest is indicated by a lt NameOfWarrantedGraph gt swp digestMethod lt digestMethod gt triple for each digest The SWP vocabulary defines URI references for identifying several digest methods These URI references are explained in table CHAPTER 1 THE SWP VOCABULARY 9 URI Reference Description swp JjcC14N shal xor Digest method combining Carroll s C14N algorithm the SHA1 digest function and the XOR aggregation function swp JjcC14N md5 xor Digest method combining Carroll s C14N algorithm the MD5 digest function and XOR as aggregation function swp SaKaDig shal xor Digest method proposed by Craig Sayer and Alan Karp using the SHA1 hash function and XOR as aggregation function Table 1 4 URI references for identifying RDF digest methods After adding the swp digest and swp digestMethod triples the warrant graph is signed as described above For verifying the integrity of the war ranted graphs an information consumer first verifies the signature of the war rant graph Afterwards the information consumer recalculates the digests of the warranted graphs using the indicated digest methods The integrity of the warrented graphs is ensured if the recalculated values equal the values given b
7. 14N shal xor dsa for instance indicates that a signature is formed by transforming the RDF graph of a named graph using Carroll s canonicalization algorithm Car03 digesting the canonical serialization and the graph name using the SHA1 hash function combining both di gests using XOR as aggregation function and finality signing the digest using the DSA signature function For verifying the signature of a named graph the information consumer requires the public key or digital certificate of the information provider The SWP vocabulary defines terms for adding public keys and certificates to published information A SWPAuthority may have a swp hasKey property The value of this property is some kind of public key which be longs to the authority For representing keys the SWP vocabulary reuses the keyInfo data structure from the XML signature recommendation ERS06 CHAPTER 1 THE SWP VOCABULARY 8 1 prefix swp lt http www w3 org 2004 03 trix swp 2 gt 2 prefix xsd lt http www w3 org 2001 XMLSchema gt 3 prefix ex lt http www fu berlin suhl bizer exampleDataset gt 4 5 ex SignedGraph 6 lt urn ISIN US4581401001 gt ex rating ex Sell ts ex SignedGraph swp assertedBy ex SignedGraph 8 ex SignedGraph swp authority lt mailto reynolds ft com gt 9 ex SignedGraph swp valid from 2005 11 20 xsd date 10 ex SignedGraph swp valid until 2005 11 30 xsd date 11 ex SignedGraph swp signatureMeth
8. FREIE UNIVERSITAT BERLIN Semantic Web Publishing Vocabulary SWP User Manual Chris Bizer chris bizer de November 2006 Contents 1 The Semantic Web Publishing Vocabulary 1 1 Authorizing Named Graphs 1 2 Signing Named Graphs 004 13 Related Work 13 1 Dublin Core Element Set 1 3 2 XML Signature DU 0 e e 2 Use Case Financial Information Integration 2 1 Domain Model 2 2 Example Data Set Bibliography 12 12 15 15 Chapter 1 The Semantic Web Publishing Vocabulary Graph names provide the hooks for asserting meta information about dis tinct graphs In order to use the graph naming mechanism for represent ing information together with quality related meta information the Named Graphs data model has to be supplemented with vocabularies for expressing the types of meta information that are relevant for an application domain One type of meta information which is commonly required in the context of Web based information systems is provenance information about the origin of information e g who said what and when This section will introduce the Semantic Web Publishing Vocabulary SWP an RDF Schema vocabulary for expressing information provision re lated meta information and for assuring the origin of information with digital signatures The vocabulary has been developed together with Jeremy Carroll Hewlett Packard Labs United Kingdom Patri
9. and Libby Miller FOAF Vocabulary Specification http xmIns com foaf 0 1 2004 retrieved 09 25 2006 Car03 Jeremy Carroll Signing RDF Graphs In Proceedings of the 2nd International Semantic Web Conference pages 369 384 2003 CBHS05 Jeremy Carroll Christian Bizer Patrick Hayes and Patrick Stickler Named Graphs Journal of Web Semantics 3 4 247 267 2005 Dun06 Dun amp Bradstreet D amp B D U N S Numbers http www dnb com US duns_update 2006 retrieved 09 25 2006 ERS06 Donald Eastlake Joseph Reagle and David Solo XML Signature Syntax and Processing RFC 3275 W3C Recommendation http www w3 org TR xmldsig core 2002 retrieved 09 25 2006 FB06 N Freed and N Borenstein RFC 2045 Multipurpose Internet Mail Extensions MIME Part One Format of Internet Message Bodies 1996 retrieved 09 25 2006 FIP95a FIPS PUB 180 1 Secure Hash Standard National Institute of Standards and Technology 1995 FIP95b FIPS PUB 186 2 Digital Signature Standard DSS National Institute of Standards and Technology 1995 HPFS06 R Housley W Polk W Ford and D Solo RFC 3280 Internet X 509 Public Key Infrastructure Certificate and Certificate 18 BIBLIOGRAPHY 19 18097 18003 ISO06 KC06 KS06 Moa06 Revocation List CRL Profile http www ietf org rfc rfc3280 txt SaaS txt 2002 retrieved 09 25 2006 ISO 3166 1 1997 Codes for the representation of names
10. ce http help yahoo com help us fin research research 34 html 2006 retrieved 09 25 2006
11. ck Stickler Nokia Finland and Pat Hayes Institute for Human and Machine Cognition United States The Semantic Web Publishing vocabulary is designed for information syndication processes in which information is passed through multiple inter mediaries These syndication processes imply three basic roles Information Providers publish information in various forms Information providers have different degrees of commitment towards published in formation e g they might believe information to be true or might be in doubt about the reliability of published information In order to prove the origin of information and to ensure that information is not altered in the syndication process information providers can digitally sign information Information Syndicators are intermediaries who collect information from CHAPTER 1 THE SWP VOCABULARY 2 xsd base64Binary xsd base64Binary swp digest swp signature 9 9 swp digestMethod swp signatureMethod swp SignatureMethod swp assertedBy or swp quotedBy swp authority 1 swp certificate xsd base64Binary swp DigestMethod Figure 1 1 The Semantic Web Publishing Vocabulary SWP multiple information providers and distribute collected information to information consumers or other syndicators Information syndicators might add meta information about the syndication process to syndi cated information They are no
12. e both digests afterwards using an aggregation function like XOR The SWP vocabulary provides terms for describing which combina tion of canonicalization digest aggregation and signature functions is CHAPTER 1 THE SWP VOCABULARY T URI Description swp JjcC14N shal xor dsa Signature method combining Carroll s C14N algorithm with the SHA1 digest function XOR as aggregation function and the DSA signature function swp JjcC14N shal xor rsa Signature method combining Carroll s C14N algorithm with the SHA1 digest function XOR as aggregation function and the RSA signature function swp JjcC14N md5 xor dsa Signature method combining Carroll s C14N algorithm with the MD5 digest function XOR as aggregation function and the DSA signature function swp SaKaDig shal xor dsa Signature method combining Sayer amp Karp s digest algorithm using the SHA1 hash function and the XOR aggregation function with the DSA signature function swp SaKaDig shal xor rsa Signature method combining Sayer amp Karp s digest algorithm using the SHA1 hash and the XOR aggregation function with the RSA signature function Table 1 3 URI references for identifying RDF signature methods used to compute a signature Each of these szgnature methods is iden tified by a URI reference Table summarizes the signature meth ods that are defined by the SWP vocabulary The signature method swp swp JjcC
13. efix dc lt http purl org dc elements 1 1 gt prefix foaf lt http xmlns com foaf 0 1 gt prefix swp lt http www w3 org 2004 03 trix swp 2 gt prefix iso lt http www daml org 2001 09 countries iso 3166 ont gt prefix fin lt http www fu berlin suhl bizer 2006 FinVoc gt prefix fd lt http www fu berlin suhl bizer exampleDataset gt fd GraphFromIntel lt urn x DUNS 047897855 gt rdf type fin Corporation rdfs label Intel Inc xsd string fin country iso US foaf homepage lt http www intel com gt lt urn x 1SIN US4581401001 gt rdf type fin Share fin emitter lt urn x DUNS 047897855 gt fd GraphfromSiemens lt urn x DUNS 316067164 gt rdf type fin Corporation rdfs label Siemens AG xsd string fin country iso DE foaf homepage lt http www siemens com gt lt urn x ISIN DE0007236101 gt rdf type fin Share fin emitter lt urn x DUNS 316067164 gt fd GraphFromPeterSmith lt urn x ISIN DE0007236101 gt fin positiveAnalystReport As Siemens agrees partnership with Novell unit SUSE EN lt urn x ISIN US4581401001 gt fin negativeAnalystReport Chiphersteller Intel will nach Firmenangaben mit milliardenschweren DE f d GraphFromJohnReynolds lt urn x DUNS 316067164 gt fin news Siemens AG currently has no plans to sell its troubled mobile phone division EN lt urn x DUNS 047897855 gt fin news Intel has record quarterly sa
14. en an authority and an information resource using warrants as an additional level of indirection This reification of the relationship allows the relationship to be described using additional proper ties such as validity and expiry date CHAPTER 1 THE SWP VOCABULARY 11 1 3 2 XML Signature Syntax The W3C XML Signature Syntax and Processing recommendation ERS06 defines a vocabulary for describing the process of computing a digital sig nature from arbitrary URL addressable data and for representing digital signatures public keys and certificates in the form of XML elements The design of the Semantic Web Publishing vocabulary was inspired by XML sig nature They main difference between both approaches is that the Semantic Web Publishing vocabulary represents signatures as RDF which facilitates the processing of SWP signatures within RDF oriented applications Second XML signature provides terms for identifying XML canonicalization methods but does not define terms or identifying RDF specific canonicalization and digest methods The SWP vocabulary closes this gap by defining terms for identifying RDF specific methods Chapter 2 Use Case Financial Information Integration Financial information portals like Wallstreet Journal Onlind Bloomberg Yahoo Financd and Google Financ enable investors to access a multitude of financial news analyst reports and postings from investment related dis cussion forums This chapter demonstrates
15. espace Moa06 The name and a description of a se curity is represented using rdfs label and dc description properties News analyst reports and discussion forum postings about a se curity are represented using the fin news fin positiveAnalystReport fin neutralAnalystReport fin negativeAnalystReport and fin posting properties Securities are related to the emitting organization by the fin emitter property Organizations are represented as instances of the class fin Organization and its subclasses fin Corporation fin AnalystHouse and fin NewsSource Organizations are identified using the Data Uni versal Numbering System D U N S which is also mapped into the URN namespace The name and the profile of an organization is represented using rdfs label and dc description properties The foaf homepage property contains a link to the main website of the organization A ISO 3166 1 country code may be assigned to an organization using the fin country property News and discussion forum postings about organizations are represented using the fin news and fin posting properties Persons are represented as instances of the class foaf Person Persons are identified by their email address The affiliation of a person to an fin Organization is expressed using the fin affiliation prop erty Persons may rate other persons analysts analyst houses and news sources A rating represents the subjective overall impres sion of the rater about the quality
16. h2 swp assertedBy ex Graph2 14 ex Graph2 swp authority lt mailto richard miller com gt 15 16 ex Graph3 17 ex Graphi swp quotedBy ex Graph3 18 ex Graphi swp sourceURL lt http www moodys com rec45 rdf gt 19 ex Graph2 swp quotedBy ex Graph3 20 ex Graph2 swp sourceURL lt http www finBlog com post32 rdf gt Pale ex Graph3 swp assertedBy ex Graph3 22 ex Graph3 swp authority lt http information syndicator com gt 23 F Figure 1 2 Graph set using the Semantic Web Publishing vocabulary for representing authorizing relationships syndicator lt http information syndicator com gt 1 2 Signing Named Graphs In order to prove the origin of information and to ensure that information is not altered in the syndication process information providers may decide to digitally sign named graphs A digital signature of a named graph is formed by computing a digest of the named graph and by signing this digest using a standard signature algorithm like DSA or RSA KS06 Graph signatures are verified by recomputing the digest of the signed graph and by decoding the original digest from the signature using the public key of the information provider If both digests are equal it is proved that the graph originates from the holder of the public key and that it has not been altered in the syndication process Content syndication processes where information is passed over multiple information syndicators pose special re
17. hFromPeterSmith contains analyst reports about the stocks of both companies The reports have been authored by Peter Smith an analyst working for Deutsche Bank The graph fd GraphFromMarkScott con tains two discussion forum posting authored by Mark Scott The graph also contains information quality ratings for different sources Mark Scott rates Peter Smith and Deutsche Bank positive and rates Dave Reynolds negative The graph fd BackgroundInformation contains background information about Peter Smith Dave Reynolds and Mark Scott The first 5 triples of the graph assert that Peter Smith is a German analyst who works for Deutsche Bank and has a StarMine performance benchmark of 88 The next 4 triples contain background information about Deutsche Bank fd GraphFromAggregator con tains provenance information about the other graphs The first 3 triples for instance represent the information that fd GraphFromIntel was asserted by Intel on 21st October 2005 Chapter will describe how background and provenance information from the last two graphs can be used by different filtering policies CHAPTER 2 USE CASE 16 oono APUNE e e e oO 9701 Hs bs As ss A SS BWWWWWWWWWWNHNNNNNNNNNKRPKRBRPBHBHB EB ROO XJDOARFRONROOOAOZJAANAAINROIOAJOdNADON O0O0OJOVAOYN prefix rdf lt http www w3 org 1999 02 22 rdf syntax ns gt prefix rdfs lt http www w3 org 2000 01 rdf schema gt prefix xsd lt http www w3 org 2001 XMLSchema gt pr
18. les Intel the world s largest chipmaker EN fd GraphFromMarkScott 4 lt urn x ISIN DE0007236101 gt fin posting I think the time OEN lt urn x 1SIN US4581401001 gt fin posting As we have seen in EN lt mailto mark scott com gt fin positiveRating lt urn x DUNS 332907323 gt lt mailto mark scott com gt fin positiveRating lt mailto peterSmith deutsche bank de gt lt mailto mark scott com gt fin negativeRating lt mailto reynolds ft com gt Figure 2 2 Example Graph Set Part 1 CHAPTER 2 USE CASE 17 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 TAs 72 73 74 75 76 TT 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 fd BackgroundInformation lt mailto peterSmith deutsche bank de gt rdf type fin Analyst foaf name Peter Smith xsd string fin country iso DE fin affiliation lt urn x DUNS 332907323 gt fin benchmark 83 xsd integer lt urn x DUNS 332907323 gt rdf type fin AnalystHouse foaf name Deutsche Bank xsd string foaf homepage lt http www db com gt fin country iso DE lt mailto reynolds ft com gt rdf type fin Analyst foaf name John Reynolds xsd string fin country iso UK fin affiliation lt urn x DUNS 42307553 gt lt urn x DUNS 42307553 gt rdf type fin NewsSource foaf name Financial Times xsd string foaf homepage lt
19. od swp JjcC14N md5 xor rsa 12 ex SignedGraph swp signature 13 AZ8QWEJO5HaDsh4iHYmsJfMDV1 xsd base64Binary 14 Figure 1 3 Signing a single graph Within RDF the keyInfo element is represented as an XML literal KC06 The swp certificate property is used to represent digital certificates of an SWPAuthority The value of the swp certificate property is the base64 encod ing of a binary ASN 1 DER X 509 certificate containing the public key of the authority Figure 1 3 shows how the Semantic Web Publishing vocabulary is used to represent the signature of a named graph The named graph ex SignedGraph is signed by the authority lt mailto reynolds ft com gt Line 11 asserts that the signature method swp JjcC14N shal xor rsa was used to calculate the signa ture given in line 13 Note that the swp signatureMethod triple is added to the graph before the digest calculation in order to be able to detect subse quently changed signature methods The swp signature triple is added after the digest calculation Before verifying the signature this triple has to be removed from the graph again Beside of signing single graphs the SWP vocabulary also provides for signing warrant graphs which assert or quote multiple graphs All graphs that are asserted or quoted by a warrant graph will be called warranted graphs in the following For ensuring the integrity of these warranted graphs SWP uses a similar technique as XML signature ERS
20. of countries and their subdivisions Part 1 Country codes International Organization for Standardization 1997 ISO 15836 2003 Information and Documentation The Dublin Core Metadata Element Set International Organization for Standardization 2003 ISO 6166 2001 Securities and related Financial Instruments International Securities Identification Numbering System ISIN 2001 retrieved 09 25 2006 Graham Klyne and Jeremy Carroll Resource Description Framework RDF Concepts and Abstract Syntax W3C Recommendation http www w3 org TR rdf concepts 2004 retrieved 09 25 2006 B Kaliski and J Staddon RFC2437 PKCS1 RSA Cryptography Specifications Version 2 0 http www ietf org rfc rfc2437 txt 1998 retrieved 09 25 2006 R Moats RFC 2141 URN Syntax http tools ietf org html rfc 2141 1997 retrieved 09 25 2006 NPJNO6 Mikael Nilsson Andy Powell Pete Johnston and Ambj rn Riv06 SK04 Yah06 Naeve Expressing Dublin Core metadata using the Resource Description Framework RDF Dublin Core Working Draft http dublincore org documents dc rdf 2006 retrieved 09 25 2006 Ronald Rivest RFC 1321 The MD5 Message Digest Algorithm http tools ietf org html rfc1321 1992 retrieved 09 25 2006 Craig Sayers and Alan Karp Computing the Digest of an RDF Graph Technical Report HPL 2003 235 Hewlett Packard Laboratories 2004 Yahoo Finance Analyst Recommendation Performan
21. of information that is published by the rated information provider Ratings are expressed using the fin positiveRating fin neutralRating and fin negativeRating proper ties The class fin Analyst is a subclass of foaf Person It consists of all persons working as analysts for a fin AnalystHouse fin Analysts may have a fin benchmark property which measures the accuracy of the recommendations of an analyst relative to his peers The score is calculated using the StarMine algorithm Yah06 Named Graphs are typed as instances of the class rdfg Graph CBHS05 CHAPTER 2 USE CASE 15 Provenance information about a graph is expressed using the Semantic Web Publishing vcabulary described in section 2 2 Example Data Set The domain model has been instantiated with example data from sev eral publicly accessible financial information portals The complete data set is available at http www wiwiss fu berlin de suhl bizer wiqa finUseCase The data was complemented with random ratings for analysts analyst houses and news sources which will be used in the next part of this thesis for demonstrating rating based filtering policies Figure 2 2 and 2 3 show the TriG serialization of a subset of the collected data The subset consists of seven named graphs Each graph contains infor mation from a specific information provider The graphs fd GraphFromIntel and fd GraphFromSiemens contain general information about Intel and Siemens The graph fd Grap
22. quirements on digital signatures for CHAPTER 1 THE SWP VOCABULARY 5 named graphs 1 Information syndicators combine graphs from different sources into graph sets and may forward these graph sets using a different seri alization syntaxes as the original documents containing the graphs Therefore a graph signature should still be verifiable if a graph is con tained in a different graph set and if this graph set is serialized using a different syntax 2 The RDF data model does not assign any semantic to blank node identifiers KC06 Two graphs that do not share blank node identi fiers may still be semantically equivalent within an application context Therefore graph signatures should be independent from blank nodes identifiers and should still be verifiable if blank nodes are renamed The Semantic Web Publishing vocabulary provides terms for represent ing digital signatures for indicating the signature method that was used to compute a signature and for representing cryptographic keys and cer tificates Table gives an overview about the signature related terms of the SWP vocabulary Graph signatures are attached to warrants using the swp signature property The value of the swp signature property is an RDF literal representing the signature of the graph that is asserted or quoted by the warrant The signature is encoded using the base64 algorithm FBO6 The swp signatureMethod property identifies the signature method that was used
23. t committed to the truth of information as they are merely quoting other sources Information consumers receive information directly from information providers or through information syndicators For assessing the quality of received information information consumers require meta information about the origin of information and the syndication process In order to verify the origin of information information con sumers might require information to be digitally signed Figure gives an overview about the Semantic Web Publishing vo cabulary The vocabulary consists of two parts The first part defines terms for authorizing information and for representing information pro vision related meta information The second part defines terms for rep resenting digital signatures The namespace of the SWP vocabulary is http www w3 org 2004 03 trix swp 2 In the following the SWP namespace is abbreviated with the prefix swp 1 1 Authorizing Named Graphs The basic idea of the SWP vocabulary is to record the authorizing relation ship between a named graph and an authority in the form of a warrant An authorizing relationship means that the authority in some sense commits it self to the content of the graph The SWP vocabulary provides terms for representing different propositional attitudes such as asserting or quoting towards a graph Warrants may also record other properties of an authorizing relationship such as the validity or expiry date
24. t have a unique authority Intuitively swp assertedBy means that the warrant records an endorsement or assertion that the graph is true while swp quotedBy means that the graph is being presented without any comment being made on its truth Figure shows an example graph set which uses the SWP vocabu lary for representing authorizing relationships The first graph ex Graph1 contains a triple recommending to sell the stock with the identifier lt urn x ISIN US4581401001 gt line 5 The graph is asserted by a warrant with the authority lt mailto reynolds ft com gt lines 6 7 Lines 8 and 9 define the validity period of the warrant ex Graphi is a self asserting graph as line 6 contains the triple ex Graph1 swp assertedBy ex Graph1 Self asserting graphs will be called warrant graphs in the following The third graph lines 16 22 quotes ex Graphi and ex Graph2 The graph is asserted by the information CHAPTER 1 THE SWP VOCABULARY 4 1 Oprefix swp lt http www w3 org 2004 03 trix swp 2 gt 2 prefix ex lt http www fu berlin suhl bizer exampleDataset gt 3 4 ex Graphi 5 lt urn ISIN US4581401001 gt ex rating ex Sell 6 ex Graphi swp assertedBy ex Graphi Ta ex Graphi swp authority lt mailto reynolds ft com gt 8 ex Graph1 swp valid from 2005 11 20 xsd date 9 ex Graphi swp valid until 2005 11 30 xsd date 10 11 ex Graph2 12 lt urn ISIN US4581401001 gt ex rating ex Buy 13 ex Grap
25. to calculate the signature Computing a digital signature for a large amount of data is usually ex pensive Therefore it is common practice to calculate a digest of the data and to sign this digest instead of the original data FIP95b There have been two proposals for computing syntax and blank node identifier independent digests of RDF graphs e Jeremy Carroll proposes an algorithm for transforming semantically equivalent graphs into a canonical serialization Car03 The algorithm renames blank nodes in a uniform fashion and sorts triples into a lex ical order After canonicalizing a graph its digest can be computed from the canonical serialization using a standard hash function like SHA1 FIP95al or MD5 Rivo06l e Craig Sayers and Alan Karp propose a second algorithm for computing digests of RDF graphs The algorithm does not rely on an inter mediate canonical serialization but computes separate hash values from each triple in the graph These values are aggregated into a set hash CHAPTER 1 THE SWP VOCABULARY 6 Property Description swp signature The value of this property is the signature to be used to authenticate the graphs with which the subject warrant is associated swp signatureMethod The value of this property is the signature method by which the signature specified for the subject warrant was constructed swp digest The value of this property contains a digest value for the subject graph
26. y the swp digest triples Figure shows a graph set consisting of two warranted graphs ex WarrantedGraphi and ex WarrantedGraph2 and a warrant graph ex WarrantGraph Line 12 contains the information that ex WarrantedGraph1 is asserted by ex WarrantGraph Line 14 and 15 contain the digest for ex WarrantedGraphi The method that was used to compute the digest is indicated in line 13 1 3 Related Work This section compares the Semantic Web Publishing Vocabulary with two related standards The Dublin Core Element Set 15003 and XML Signature Syntax ERSOG6 1 3 1 Dublin Core Element Set Similar to SWP warrants the Dublin Core SO03 elements dc creator dc publisher and dc contributor relate information resources to persons or institutions The difference between both vocabularies lies in types of repre sentable relations The Semantic Web Publishing vocabulary is focused on the commitment of an authority towards the truth of information Asserting CHAPTER 1 THE SWP VOCABULARY 10 1 prefix swp lt http www w3 org 2004 03 trix swp 2 gt 2 prefix xsd lt http www w3 org 2001 XMLSchema gt 3 prefix ex lt http www fu berlin suhl bizer exampleDataset gt 4 5 ex WarrantedGraphi 6 lt urn ISIN US4581401001 gt ex rating ex Sell 7 y 8 ex WarrantedGraph2 9 lt urn ISIN DE0007236102 gt ex rating ex Buy 10 11 ex WarrantGraph 12 ex WarrantedGraphi swp assertedBy ex WarrantGraph 13
27. ystReport rdfs Literal fin posting rdfs Literal fin Bond fin Certificate fin Fund emits fin Organization fin AnalystHouse rdfs label rdfs Literal de description rdfs Literal foaf homepage foaf Document fin duns rdfs Literal 0 fin country iso Country fin news rdfs Literal fin posting rdfs Literal fin NewsSource NN YN fin Corporation foaf Person affiliation foaf name rdfs Literal de description rdfs Literal foaf mbox rdfs Resource foaf homepage foaf Document fin affiliation fin Company fin country iso Country fin positiveRating rdfs Resource fin neutralRating rdfs Resource fin negativeRating rdfs Resource fin Analyst fin benchmark rdf Literal rdfg Graph lt asserts rdfs label rdfs Literal swp assertedBy swp Warrant swp quotedBy swp Warrant swp authority swp Authority Figure 2 1 UML class diagram of the domain model CHAPTER 2 USE CASE 14 ated using the fin prefix Figure 2 I gives an overview about the classes of the model Securities are represented as instances of the class fin Security and its subclasses fin Share fin Bond fin Fund fin Certificate Securities are identified by their International Securities Identifying Number ISIN which is mapped into the Universal Resource Name URN nam
Download Pdf Manuals
Related Search