Technologies for semi automatic metadata creation

1
Technologies for (semi-) automatic metadata
creation http//gate.ac.uk/
http//nlp.shef.ac.uk/ Diana Maynard University
of Sheffield KnowledgeWeb WP 1.3 meeting, Crete,
14 May 2004
2
Overview

• USFD is mainly concerned in this WP with best
  practices and guidelines for ontology-based web
  applications
• State-of-the-art systems and platforms for
  metadata creation
• Metadata is created through semantic tagging
• Metadata can be represented as inline
  (modification of the original document) or
  standoff (separate storage from the document)

3
Semi-automatic v automatic metadata creation

• Semi-automatic methods are more reliable, but
  require human intervention
• MnM requires initial human annotation
  pre-defined ontology
• S-CREAM
• AERODAML
• Automatic methods less reliable, but suitable for
  large volumes of text, and offer a dynamic view
• SemTag semantic tagging from ontology
• KIM semantic tagging and ontology population
• hTechSight semantic tagging, ontology population
  and evolution

4
Semi-automatic methods

• MnM
• S-CREAM

5
MnM

• Semi-automatic in that it requires initial
  training by user
• Uses pre-defined set of concepts in ontology
• User browses web and manually annotates his
  chosen pages
• System learns annotation rules, tests them, and
  takes over annotation, populating ontologies with
  the instances found
• Precision and recall are not perfect, however
  retraining is possible at any stage

6
S-CREAM

• Semi-automatic CREAtion of Metadata
• Uses Onto-O-Mat Amilcare
• Trainable for different domains
• Aligns conceptual markup (which defines
  relational metadata) provided by e.g. Ont-O-Mat
  with semantic markup provided by Amilcare

7
Annotated data in S-CREAM
8
Amilcare

• Amilcare learns IE rules from pre-annotated data
  (e.g. using Ont-O-Mat)
• Uses GATE (ANNIE) for pre-processing applies
  rules learnt in training phase to new documents
• Concepts need to be pre-defined, but system can
  be trained for new domain
• Can be tuned towards precision or recall

9
Automatic methods

• SemTag
• KIM
• h-Techsight

10
SemTag and KIM

• SemTag and KIM both annotate webpages using
  instances from an ontology
• Main problem is to disambiguate such instances
  which occur in multiple parts of the ontology
• SemTag aims for accuracy of classification,
  whereas KIM aims more for recall (finding all
  instances)
• KIM also uses IE to find new instances not
  present in ontology

11
SemTag

• Automated semantic tagging of large corpora,
  using TAP ontology (contains 65K instances)
• Largest scale semantic tagging effort to date
• Uses concept of Semantic Label Bureau
• Annotations are stored separately from web pages
  (standoff markup)
• Uses corpus-wide statistics to improve quality of
  tagging, e.g. automated alias discovery
• Tags can be extracted using a variety of
  mechanisms, e.g. search for all tags matching a
  particular object

12
SemTag Architecture
13
KIM

• Uses an ontology (KIMO) with 86K/200K instances
• Lookup phase marks instances from the ontology
• High ambiguity of instances with the same label
  (e.g. locations belonging to different countries)
• Disambiguation uses an Entity Ranking algorithm,
  i.e., priority ordering of entities with the same
  label based on corpus statistics
• Lookup is combined with rule-based IE system
  (from GATE) to recognise new instances of
  concepts and relations
• Special KB enrichment stage where some of these
  new instances are added to the KB

14
KIM (2)
15
h-TechSight KMP

• Knowledge management platform for fully automatic
  metadata creation and ontology population, and
  semi-automatic ontology evolution, powered by
  GATE and ToolBox.
• Data-driven analysis of ontologies enables trends
  of instances to be monitored
• Uses GATE to support the instance-based evolution
  of ontologies in the Chemical Engineering domain.
• Analysis of unrestricted text to extract
  instances of concepts from such ontologies
• Instances populated into a domain-specific
  ontology and/or exported to an Access / Oracle
  database

16
1
2
3
4
17
Ontology-based IE in h-TechSight

• Ontology-Based IE for semantic tagging of job
  adverts, news and reports in chemical engineering
  domain
• Semantic tagging used as input for ontological
  analysis
• Fundamental to the application is a
  domain-specific ontology
• Terminological gazetteer lists are linked to
  classes in the ontology
• Rules classify the mentions in the text wrt the
  domain ontology
• Annotations output into a database or as an
  ontology

18
Limitations

• h-Techsight uses rule-based IE system
• Requires human expert to write rules
• Accurate on restricted domains with small
  ontologies
• Adaptation to a new domain / ontology may require
  some effort

19
Summary

• Tradeoff between semi-automatic and fully
  automatic systems, dependent on application,
  corpus size etc
• Tradeoff between rule-based and ML techniques for
  IE
• Tradeoff between dynamic vs static systems