The artof of knowledge engineering, or: knowledge engineering of art
1. The Art of Knowledge Engineering
or: knowledge engineering of art
Guus Schreiber
2. Knowledge engineering:
From art to discipline
â Objective of knowledge engineering (KE) in 90âs
â Use of knowledge patterns for âexpertâ tasks
3. The popularity of âontologyâ
â One of the first ontologies: traffic ontology
â âOn representational promiscuityâ
â âVacuous paper with no contentâ (Brian Gaines, KAW 93)
â Now seen as panacea for the Holy Grail of
information integration
â There is even a Web language for it
â But will it stand the test of time?
â are our current conceptions of formal classes and
properties sufficient to grasp the complexity of the Web?
4. The Web: knowledge engineering
for the masses
â KE is outside the former small research community
â Everyone is building hierarchies and describes classes
5. Intermezzo:
Knowledge democracy
â From a human-rights point of view the Web is a leap
forward
â Possibility for âeveryoneâ to access knowledge
â the empowered citizen
â But history will tell whether this remains true
6. Web KE builds on a long tradition of
vocabulary construction
9. The popularity of
âontologyâ alignment
â Creating the (missing) links the the Linked Open Data
Cloud
â Multitude of alignment techniques available
â Large evaluation initiative
â OAEI
â But will our alignment methods stand the test of
time? Are the results good enough?
10. Semantic search types
â DISAMBIGUATE: Can you give me alternative
interpretations of term T?
â DESCRIBE: Can you give me more information about
concept C or Individual I?
â QUESTION ANSWERING: does property P hold for
object O?
â ANSWER QUESTIONING: Jeopardy!
â RELATION SEARCH: in what way(s) are object O1 and
O2 related?
13. How are Picasso and Matisse
connected: Georges Braque
Style- and time-based
not trivial
14. How are Picasso and Matisse
connected: 1907
the changes in the art world in Paris anno 1907
difficult
15. Matisse and âles fauvesâ
Where does this term come from?
1905: the story of the critic
difficult
16. The importance of narratives
â Users like to get the âstoriesâ behind the navigation
paths in the graph
â For this we need to have some minimal
âunderstandingâ of the meaning of the paths
â Well-constructed minimal ontologies can provide
such interpretations
â graph patterns
â Example: an event ontology
18. Problems in ontology alignment
â We have not agreed on an adequate alignment
vocabulary
â misuse of owl:sameAs
â We have no adequate methodology for evaluating
alignments (Tordai et al., 2011)
â In particular, people do not agree on how different
classes align
â and this is not because they donât do it ârightâ
19. Beyond categories
â The set theory on which ontology languages are built
is inadequate for modelling how people think about
categories (Lakoff)
â Category boundaries are not hard: cf. art styles
â People think of prototypes; some examples are very
prototypical, others less
â We also need to make meta-distinctions explicit
â organizing class: âfurnitureâ
â base-level class: âchairâ
â domain-specific: âWindsor chairâ
20. KE for the Web: the way forward?!
â We are only scratching the surface in semantic
search
â large-scale experimentation needed
â small minimal ontologies acting as search patterns
â We need a revised alignment vocabulary
â taking Lakoffâs notions into account
â Attention for semantic detail matters
â in search, in aligment, for story telling
â and lay knowledge engineers are providing it!
â Combining this with statistical technqiues is a
powerful combination