vocabulary matching for book indexing suggestion in linked libraries – a prototype implementation...
Post on 21-Dec-2015
218 views
TRANSCRIPT
![Page 1: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/1.jpg)
Vocabulary Matching for Book IndexingSuggestion in Linked Libraries
– A Prototype
Implementation & Evaluation
Antoine Isaac, Dirk Kramer, Lourens van der Meij, Shenghui Wang, Stefan Schlobach, Johan Stapel
![Page 2: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/2.jpg)
Problem: subject indexing
• Describing subjects of books• Using concepts from vocabularies (e.g. thesauri)
![Page 3: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/3.jpg)
Problem: re-indexing
• Describing a book that has already be described• With a new vocabulary
– Fitting a different context (e.g., different libraries)
![Page 4: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/4.jpg)
Why re-indexing at KB?
• The Dutch National Library (KB) holds many books that are also in other Dutch public libraries
• KB deposit uses Brinkman thesaurus for indexing• Public Libraries use Biblion thesaurus
KBDeposit
Collection
DutchPublic
Libraries
Biblion Brinkman
overlap betweenbook collections
![Page 5: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/5.jpg)
A wider issue• KB shares books with many other libraries• All having their own description practices
KB
KBDeposit
Coll.
KBScientific
Coll.
DutchPublic
Libraries
LC(US Nat.
Lib)
BnF(FrenchNat. Lib)
DNB(GermanNat. Lib)
DutchBook-trade
Biblion
NURBISACsubjectcodes
Brinkman GTT
NBCclass.
UNESCOclass.
KBCorporatie+ Persoon
RAMEAUsubject
headings
LCSHsubject
headings
DDCDewey
decimalclass.
SWDsubject
headings
Personennamendatei
LCauthority
file
AutoritésBNF
otherclassifications
domain/discipline
classifications
subjectthesauri /
subj. headinglists
bookcollectiondatasets
person/corporation
data
Doel-groep
--audience
overlap between book collections(thickness indicates degree of overlap)
Vertical adjustment between a coll. and KOSsdenotes KOSs' being used to describe that coll.
![Page 6: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/6.jpg)
Room for improvement?
• Libraries devote large resources to indexing– 20 people at KB– About 20,000 books per year
• Leveraging already existing descriptions for re-indexing can be beneficial for both sides
![Page 7: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/7.jpg)
Alignment and re-indexing
• STITCH project– Tackling semantic interoperability in Cultural Heritage– Using ontology alignment
• Mappings between concepts from different vocabularies can be used for re-indexingBasic idea: replace concepts in descriptionsby conceptually equivalent concepts
![Page 8: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/8.jpg)
Goal: a re-indexing prototype
• Past: preliminary experiments with KB data
• Now: building a prototype and– plugging it onto the KB production system– having it evaluated by its potential users (indexers)
• Prototype case: Dutch public libraries / KBSuggesting Brinkman subjects based on Biblion ones
![Page 9: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/9.jpg)
Alignment and re-indexing: requirements
Subjects can be complex
• Mappings between groups of concepts "Travel guides" + "Spain" → "Spain; travel guides"
Concepts are used in descriptions
• Mappings taking into account extensional semantics"Building engineering"
→ "Learning material ; building engineering"
![Page 10: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/10.jpg)
Obtaining re-indexing rules
• Lexical alignments are not good enough
• Probabilistic rules are calculated– Using extension of concepts: existing indexing– Simple probabilities, with adhoc adjustment
"Travel guides","Spain"→"Spain; travel guides", 0.982
• Not only based on Biblion subjects– AUT – main authors of books– KAR – “characteristic”– DGP – intellectual level/target group
![Page 11: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/11.jpg)
Demo
Doesn't work?
![Page 12: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/12.jpg)
User study
• Quantitative aspect– How well does the tool compare to human subject
indexing?
• Qualitative aspect– User satisfaction– Improvement suggestion
![Page 13: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/13.jpg)
Evaluation setting
• 6 indexers• 6 weeks• 284 books• Evaluation integrated in daily indexing work
• Pre-evaluation briefing• Questionnaire during evaluation • Post-evaluation de-briefing & questionnaire
![Page 14: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/14.jpg)
User study results
• Top ranked mappings are indeed much better
• Individual book satisfaction level > 70%
Suggestion class # suggestions precision recall
blue 308 72.7% 47.9%
purple 1,188 10.7% 27.1%
red 2,525 1.11% 5.98%
non suggested 89 19.0%
![Page 15: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/15.jpg)
User study results (1)
• But the general satisfaction is lower– Only two out of six would use the tool as such
• Quality of suggestions– Lower-level suggestions are often not meaningful
• Perception of suggestions' quality– Long lists with wrong suggestions ad the end are bad– Ranking is appreciated, but it is not enough
![Page 16: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/16.jpg)
User study results (2)
Suggestions were found promising• Bridging the indexing gap between collections
– Different indexing strategies
"Persian language" (Biblion)
vs. "Iranian language and literature" (Brinkman)
Lots of suggestions for improvement• More re-indexing!
– Suggesting concepts from other vocabularies– More context metadata as input
![Page 17: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/17.jpg)
Conclusions
• Shows the potential of re-using data in a library network
• Alignment approach fitting indexing practice
• Concrete demonstration, in KB production environment
• Technology transfer: KB wants to continue efforts
• Flexibility: architecture ready to exploit other vocabularies– Linked data & SKOS
![Page 18: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/18.jpg)
Prototype components
Sesame SKOSRDF store
STITCH script(VisualBasic)
STITCHstylesheet (XSLT)
Indexer
WinIBWcataloguing interface
IE
GGC cataloguingsystem
LOD SPARQLendpoints
suggestion service(SWI-Prolog)
vocabularyservice
(Java/Tomcat)
lexical alignmentsSesame RDF store
![Page 19: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/19.jpg)
Linked libraries?
KB
KBDeposit
Coll.
KBScientific
Coll.
DutchPublic
Libraries
LC(US Nat.
Lib)
BnF(FrenchNat. Lib)
DNB(GermanNat. Lib)
DutchBook-trade
Biblion
NURBISACsubjectcodes
Brinkman GTT
NBCclass.
UNESCOclass.
KBCorporatie+ Persoon
RAMEAUsubject
headings
LCSHsubject
headings
DDCDewey
decimalclass.
SWDsubject
headings
Personennamendatei
wikipedia.nl
wikipedia.de
LCauthority
file
AutoritésBNF
existing KOS alignment
potential KOS alignment of interest
overlap between book collections(thickness indicates degree of overlap)
otherclassifications
domain/discipline
classifications
subjectthesauri /
subj. headinglists
bookcollectiondatasets
person/corporation
data
othersLCSH
currently available entry point tothe LOD cloud
Vertical adjustment between a coll. and KOSsdenotes KOSs' being used to describe that coll.
Doel-groep
--audience
![Page 20: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/20.jpg)
Thank you!
• Questions?
![Page 21: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/21.jpg)
Screenshots
![Page 22: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/22.jpg)
WinIBW production tool
![Page 23: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/23.jpg)
STITCH suggestion tool
![Page 24: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/24.jpg)
Original metadata
![Page 25: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/25.jpg)
Concept suggestions
![Page 26: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/26.jpg)
Comparing with human re-indexing
![Page 27: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/27.jpg)
Complement: lexical alignments
![Page 28: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/28.jpg)
Adding subjects using thesaurus access
![Page 29: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/29.jpg)
Concept suggestions
![Page 30: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/30.jpg)
Saving and back to WinIBW
![Page 31: Vocabulary Matching for Book Indexing Suggestion in Linked Libraries – A Prototype Implementation & Evaluation Antoine Isaac, Dirk Kramer, Lourens van](https://reader036.vdocuments.us/reader036/viewer/2022062421/56649d625503460f94a44e63/html5/thumbnails/31.jpg)
Screenshots
• Back