Download presentation
Presentation is loading. Please wait.
1
Computing FOAF Co-reference Relations with Rules and Machine Learning Jennifer Sleeman and Tim Finin University of Maryland, Baltimore County The Third International Workshop on Social Data on the Web, November 2010 http://ebiquity.umbc.edu/paper/html/id/506/
2
FOAF Friend of a Friend (FOAF) vocabulary describes people and their relationships One of oldest and most widely used ontologies Does not include a globally unique identifier Inverse functional properties (IFPs) help Multiple foaf instances referring to the same person are common Increasingly so with more linked data introduction foaf co-reference approach methodology evaluation conclusions
3
Linking data Data integration requires linking instances from different data sets Linking foaf instances is a common and typical use case Sindice reports 23 foaf instances all referring to Sir Tim Berners Lee23 Probably more than my query revealed Only a handful are linked via owl:sameAs Automatically linking foaf instances is not always easy introduction foaf co-reference approach methodology evaluation conclusions
4
Example 1 Bijan Parsia Bijan Bijan Parsia Parsia http://tw.rpi.edu/wiki/Special:ExportRDF/Bijan_Parsia f49a6854842c5fa76dc0edb8e82f8fe04fd56bc9 Bijan Parsia Bijan Parsia bparsia bparsia http://trust.mindswap.org/cgi-bin/FilmTrust/foaf.cgi?user=bparsia#tt0084827-bparpia Common properties but can we say this is the same person…
5
Example 2 James A. Hendler James Hendler http://ebiquity.umbc.edu/papers/select/person/James/Hendler/ http://ebiquity.umbc.edu/person/foaf/James/A./Hendler/foaf.rdf 0b62d4242736e64be6138547c79a811b3e82fd52 Jim Hendler Jim Hendler Tetherless World Constellation Chair jhendler jhendler http://www.cs.rpi.edu/~hendler/foaf.rdf Aliases and slight name variations…
6
Example 3 08445a31a78661b5c746feff39a9db6e4e2cc5cf David Wood dw2 http://identi.ca/dw2/foaf David Wood Dr. David Wood prototypo 37c8d030d4e615d05f31625b3460532a3f4e214e piprototypo http://www.itee.uq.edu.au/~dwood/dave.rdf#me What if mbox_sha1sums are different?
7
Example 3 cont. David Wood http://www.mindswap.org/2004/owl/mindswappers#David.Wood Which David Wood was a mindswapper?
8
Example 5 08445a31a78661b5c746feff39a9db6e4e2cc5cf jgolbeck jgolbeck http://trust.mindswap.org/cgi-bin/FilmTrust/foaf.cgi?user=jgolbeck Jennifer Golbeck Jennifer Jennifer Golbeck Golbeck http://tw.rpi.edu/wiki/Special:ExportRDF/Jennifer_Golbeck Could jgolbeck and Jennifer Golbeck be the same person …
9
Example 5 cont. Jennifer Golbeck http://www.cs.umd.edu/~golbeck/daml/golbeckFOAF.rdf Jennifer Golbeck http://www.mindswap.org/2004/owl/mindswappers#Jennifer.Golbeck Which profile is most recent/relevant?
10
Our Contributions Treating foaf smushing as entity co-reference Use machine learning to train a classifier for recognizing co-referent foaf instance Combine this with rule-based evidence Use of narrower RDF properties to express co- reference, avoiding overuse of owl:sameAs Use of a greedy algorithm for iteratively clustering co-referent entities and re-evaluating their potential co-reference relations introduction foaf co-reference approach methodology evaluation conclusions
11
Co-Reference in FOAF Approach problem like cross-document co- reference resolution in text Match pairs FOAF agents Use rules and properties Assign new properties to represent coref and notCoref relationships Cluster co-referent pairs introduction foaf co-reference approach methodology evaluation conclusions
12
Cross-Document Co-reference Resolution Determine when two documents mention the same entity Are two documents that talk about “George Bush” talking about the same George Bush? Is a document mentioning “Mahmoud Abbas” referring to the same person as one mentioning “Muhammed Abbas”? What about “Abu Abbas”? “Abu Mazen”? Drawing appropriate inferences from multiple documents demands cross- document co-reference resolution 2008 NIST Text Analysis Conference
13
TAC KBP: Entity Linking John Williams Richard Kaufman goes a long way back with John Williams. Trained as a classical violinist, Californian Kaufman started doing session work in the Hollywood studios in the 1970s. One of his movies was Jaws, with Williams conducting his score in recording sessions in 1975... John Williamsauthor1922-1994 J. Lloyd Williamsbotanist1854-1945 John Williamspolitician1955- John J. WilliamsUS Senator1904-1988 John WilliamsArchbishop1582-1650 John Williamscomposer1932- Jonathan Williamspoet1929- Michael Phelps Debbie Phelps, the mother of swimming star Michael Phelps, who won a record eight gold medals in Beijing, is the author of a new memoir,... Michael Phelpsswimmer1985- Michael Phelpsbiophysicist1939- Michael Phelps is the scientist most often identified as the inventor of PET, a technique that permits the imaging of biological processes in the organ systems of living individuals. Phelps has... Given an entity mention in an article, find the link to the right Wikipedia entity if one exists. 2009 NIST TAC Knowledge Base Population Track
14
Smushing Smushing is the traditional term used for recognizing that two “blank nodes” refer to the same thing and merging them Smushing Past work on smushing has exploited IFPs (e.g., foaf:mbox), heuristic similarity metrics and custom SPARQL queries owl:sameAs is often used to relate smushed nodes, enabling a reasoner to effect the merging rdf:seeAlso used to find related foaf data introduction foaf co-reference approach methodology evaluation conclusions
15
Smushing introduction foaf co-reference approach methodology evaluation conclusions foaf: Person "foo@gmail.com" rdfs:type foaf:mbox foaf:knows foaf:nick ”bar" owl:sameAs foaf:mbox
16
Smushing introduction foaf co-reference approach methodology evaluation conclusions foaf: Person "foo@gmail.com" rdfs:type foaf:knows foaf:nick ”bar" foaf:mbox
17
owl:sameAs considered harmful Known problems – Temporally qualified data (Ding vs. Ding)Ding vs. Ding – Noisy data (Clinton vs. Clinton)Clinton vs. Clinton – Referentially opaque contexts (John likes the Morning Star beautiful) Referentially opaque contexts Halpin et. Al (2010) suggest a vocabulary for similarity relations similarity.owlsimilarity.owl We use two weaker predicates: coref & notCoref – Defer the sameAs problem to applications introduction foaf co-reference approach methodology evaluation conclusions
18
Co-Reference in FOAF coref: transitive, symmetric and reflexive; has sameAs as subproperty notCoref: symmetric and irreflexive but not transitive; has differentFrom as subproperty :coref a owl:TransitiveProperty, owl:SymmetricProperty, owl:ReflexiveProperty owl:sameAs rdfs:subPropertyOf :coref. :notCoref a owl:SymmetricProperty, owl:IrreflexiveProperty. owl:differentFrom rdfs:subPropertyOf :notCoref. {?a :notCoref ?b. ?b :coref ?c.} => {?a :notCoref ?c} {?a foaf:knows ?b.} => {?a :notCoref ?b} The :coref and :notCoref properties that we use instead of owl:sameAs introduction foaf co-reference approach methodology evaluation conclusions
19
Batch Approach Given a potentially large set of foaf instances Generate candidate pairs Evaluate each pair for co-reference Using rules and classifier independently Each results in a {coref, notCoref, unknown} decision Trust rules over classifier Designate pairs as co-referent Create Clusters introduction foaf co-reference approach methodology evaluation conclusions
20
Ingest Extract triples from FOAF profiles Add each foaf agent as new entity in database Entity URLs followed in foaf:knows graph to get additional information introduction foaf co-reference approach methodology evaluation conclusions
21
Approach: System Architecture introduction foaf co-reference approach methodology evaluation conclusions ingestion candidate pair generation candidate pair generation rule-based reasoning rule-based reasoning machine learning machine learning Model Generation Abstract entity generation Potential pairs: reduces classifier workload deductive decisions deductive decisions predictions clusters form new abstract entities Co-referent designation and clustering
22
Candidate Pairs Filter pairs reduce matching set Use simple string matching predicates Dice score for 3-grams Apply both to values of common properties and also cross-property values Experiment 2 ~30% reduction Reductions vary based on data set introduction foaf co-reference approach methodology evaluation conclusions
23
Input data sources FOAF profiles extracted from Swoogle Also used URLS extracted from tests conducted in previous work Distribution of URLs from Experiment 2 introduction foaf co-reference approach methodology evaluation conclusions
24
Methodology: Rule-based Model Rules conclude that two instances are co- referent, not co-referent or draw no conclusion (the most common outcome) Basic co-reference rule: {?p a owl:IFP. ?a ?p ?x. ?b ?p ?x) => {?a :coref ?b} {?p a owl:FP. ?a ?p ?x. ?a ?p ?y.) => { ?x :coref ?y} introduction foaf co-reference approach methodology evaluation conclusions
25
Methodology: Rule-based Model In text processing, very similar name mentions in a document more likely to be co-referent It also is used in disambiguating name men- tions in citations in a single paper or Web page A similar heuristic is useful for a “knows graph” extracted from a single foaf profile {?a foaf:knows ?b. ?a foaf:knows ?c. ?b neq ?c} => {?b :notCoref ?c} introduction foaf co-reference approach methodology evaluation conclusions
26
Methodology – Vector Model Support Vector Machine linear kernel Features: – Match/nomatch of any IFPs – Distance measures over common property values (Levenshtein & 3-gram Dice score) – Alias and entity mention resolution – Property specific feature comparison – Knows graph comparisons: Jaccard coef of similarity of foaf names of one-hop neighbors introduction foaf co-reference approach methodology evaluation conclusions
27
Methodology: Clustering Pairs form clusters Clusters used as part of system evaluation Can result in: – Entity to Entity pairing – Cluster to Entity pairing – Cluster to Cluster pairing Greedy process with a confidence threshold Use rule-based model to eliminate known non-coreferent pairs introduction foaf co-reference approach methodology evaluation conclusions
28
Methodology – Clustering Instance matching can result in new cluster formation and cluster matching can result in merged clusters. introduction foaf co-reference approach methodology evaluation conclusions
29
Evaluation Two experiments – E1: 50,000 triples, over 500 entity mentions, 600 classes used for training – E2: 250,000 triples, over 3500 entity mentions, over 1800 classes for training 10-fold cross-validation tests introduction foaf co-reference approach methodology evaluation conclusions
30
Evaluation Pairs Rule Conclusion 9138326differentFromUndetermined 47184inverse functionalUndetermined 2402inverse functionalCo-referent 8687410knows graphUndetermined 9138326sameAsUndetermined 1047874knowsNot Co-referent For E1: 900 pairs non-match, majority undetermined E2: Results shown below introduction foaf co-reference approach methodology evaluation conclusions
31
Evaluation Results promising During our E2 clustering phase, the first phase 90% accuracy Second phase no new relationships among pairs, cluster to cluster pairing occurred Classification Results using 10-fold Validation introduction foaf co-reference approach methodology evaluation conclusions
32
Evaluation Retrieving additional FOAF profiles based on knows graph Quickly retrieve large number of entities Tightly linked – reduced diversity of analyzed data – more entities that are co-referent Future experiments: a diversity filter spanning domains introduction foaf co-reference approach methodology evaluation conclusions
33
Future Work Evaluating the contribution of each rule and SVM feature to performance Other ML approaches, e.g., markov logic, EM Exploiting better clustering algorithms Adding more features, e.g. non-foaf vocabu- lary, non-RDF data (e.g., hosting site) Applying approach to other RDF instances Scalability: Providing a non-batch, streaming service Offering a coref Web service introduction foaf co-reference approach methodology evaluation conclusions
34
Conclusions We can treat instance linking as co-reference resolution & exploit in-doc and xdoc distinction Good results with an ensemble approach combining rules and an SVM classifier Apply clustering to form groups of co-referent relations and reprocess Promising initial results introduction foaf co-reference approach methodology evaluation conclusions
35
http://ebiquity.umbc.edu/
36
Introduction Determine if two entities are co-referent Co-reference common in human language, database address records, and even official government records Use of name and other properties evidence of co-reference Merging information More complete representation introduction foaf co-reference approach methodology evaluation conclusions
37
Introduction Semantic Web Co-referent relations Experimented using an ensemble approach Used both RDF and OWL rules and a Support Vector Machine to classify pairs of FOAF instances Introduced a new coref and notCoref predicate to convey co-referent relationships among relations introduction foaf co-reference approach methodology evaluation conclusions
38
Example 4 Richard Stallman rms 6852748ed575e3a6d0c0de6353147f8308f72ecd http://www.advogato.org/person/rms/foaf.rdf 1f3af991f2b053b34178c093fe96725cfa569503 Richard Stallman I'm the president of the Free Software Foundation. … rms … http://identi.ca/rms/foaf mbox_sha1sums are different but this is the same person…
39
Example 6 Yarden Katz jordan milesdavis http://www.mindswap.org/~katz/foaf.rdf Jordan Katz jordan acba421f117a4b32dbb14eb1971f6a21d9f17deb http://www.advogato.org/person/jordan/foaf.rdf Names are different but nicknames are the same…
40
Methodology: input data introduction foaf co-reference approach methodology evaluation conclusions FOAF property distribution from earlier data (March 2010)
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.