Chapter 7 Ontology Engineering Grigoris Antoniou Frank van

Chapter 7 Ontology Engineering Grigoris Antoniou Frank van Harmelen 1 Chapter 7 A Semantic Web Primer

Lecture Outline 1. 2. 3. 4. 5. 6. 2 Introduction Constructing Ontologies Manually Reusing Existing Ontologies Semiautomatic Ontology Acquisition Ontology Mapping On-To-Knowledge SW Architecture Chapter 7 A Semantic Web Primer

Methodological Questions – – – l Many of these questions for the Semantic Web have been studied in other contexts – 3 How can tools and techniques best be applied? Which languages and tools should be used in which circumstances, and in which order? What about issues of quality control and resource management? E. g. software engineering, object-oriented design, and knowledge engineering Chapter 7 A Semantic Web Primer

Lecture Outline 1. 2. 3. 4. 5. 6. 4 Introduction Constructing Ontologies Manually Reusing Existing Ontologies Semiautomatic Ontology Acquisition Ontology Mapping On-To-Knowledge SW Architecture Chapter 7 A Semantic Web Primer

Main Stages in Ontology Development Determine scope 2. Consider reuse 3. Enumerate terms 4. Define taxonomy 5. Define properties 6. Define facets 7. Define instances 8. Check for anomalies Not a linear process! 1. 5 Chapter 7 A Semantic Web Primer

Determine Scope l There is no correct ontology of a specific domain – l What is included in this abstraction should be determined by – – 6 An ontology is an abstraction of a particular domain, and there always viable alternatives the use to which the ontology will be put by future extensions that are already anticipated Chapter 7 A Semantic Web Primer

Determine Scope (2) l Basic questions to be answered at this stage are: – – 7 What is the domain that the ontology will cover? For what we are going to use the ontology? For what types of questions should the ontology provide answers? Who will use and maintain the ontology? Chapter 7 A Semantic Web Primer

Consider Reuse l l With the spreading deployment of the Semantic Web, ontologies will become more widely available We rarely have to start from scratch when defining an ontology – 8 There is almost always an ontology available from a third party that provides at least a useful starting point for our own ontology Chapter 7 A Semantic Web Primer

Enumerate Terms l Write down in an unstructured list all the relevant terms that are expected to appear in the ontology – – l Traditional knowledge engineering tools (e. g. laddering and grid analysis) can be used to obtain – – 9 Nouns form the basis for class names Verbs (or verb phrases) form the basis for property names the set of terms an initial structure for these terms Chapter 7 A Semantic Web Primer

Define Taxonomy l Relevant terms must be organized in a taxonomic hierarchy – l Ensure that hierarchy is indeed a taxonomy: – 10 Opinions differ on whether it is more efficient/reliable to do this in a top-down or a bottom-up fashion If A is a subclass of B, then every instance of A must also be an instance of B (compatible with semantics of rdfs: sub. Class. Of Chapter 7 A Semantic Web Primer

Define Properties l l Often interleaved with the previous step The semantics of sub. Class. Of demands that whenever A is a subclass of B, every property statement that holds for instances of B must also apply to instances of A – 11 It makes sense to attach properties to the highest class in the hierarchy to which they apply Chapter 7 A Semantic Web Primer

Define Properties (2) l l While attaching properties to classes, it makes sense to immediately provide statements about the domain and range of these properties There is a methodological tension here between generality and specificity: – – 12 Flexibility (inheritance to subclasses) Detection of inconsistencies and misconceptions Chapter 7 A Semantic Web Primer

Define Facets: From RDFS to OWL l l Cardinality restrictions Required values – – – l Relational characteristics – 13 owl: has. Value owl: all. Values. From owl: some. Values. From symmetry, transitivity, inverse properties, functional values Chapter 7 A Semantic Web Primer

Define Instances l l l Filling the ontologies with such instances is a separate step Number of instances >> number of classes Thus populating an ontology with instances is not done manually – – 14 Retrieved from legacy data sources (DBs) Extracted automatically from a text corpus Chapter 7 A Semantic Web Primer

Check for Anomalies l An important advantage of the use of OWL over RDF Schema is the possibility to detect inconsistencies – l Examples of common inconsistencies – – – 15 In ontology or ontology+instances incompatible domain and range definitions for transitive, symmetric, or inverse properties cardinality properties requirements on property values can conflict with domain and range restrictions Chapter 7 A Semantic Web Primer

Lecture Outline 1. 2. 3. 4. 5. 6. 16 Introduction Constructing Ontologies Manually Reusing Existing Ontologies Semiautomatic Ontology Acquisition Ontology Mapping On-To-Knowledge SW Architecture Chapter 7 A Semantic Web Primer

Existing Domain-Specific Ontologies l l Medical domain: Cancer ontology from the National Cancer Institute in the United States Cultural domain: – – – l 17 Art and Architecture Thesaurus (AAT) with 125, 000 terms in the cultural domain Union List of Artist Names (ULAN), with 220, 000 entries on artists Iconclass vocabulary of 28, 000 terms for describing cultural images Geographical domain: Getty Thesaurus of Geographic Names (TGN), containing over 1 million entries Chapter 7 A Semantic Web Primer

Integrated Vocabularies l l Merge independently developed vocabularies into a single large resource E. g. Unified Medical Language System integrating 100 biomedical vocabularies – l The semantics of a resource that integrates many independently developed vocabularies is rather low – 18 The UMLS metathesaurus contains 750, 000 concepts, with over 10 million links between them But very useful in many applications as starting point Chapter 7 A Semantic Web Primer

Upper-Level Ontologies l Some attempts have been made to define very generally applicable ontologies – l l 19 Mot domain-specific Cyc, with 60, 000 assertions on 6, 000 concepts Standard Upperlevel Ontology (SUO) Chapter 7 A Semantic Web Primer

Topic Hierarchies l Some “ontologies” do not deserve this name: – l l l 20 simply sets of terms, loosely organized in a hierarchy This hierarchy is typically not a strict taxonomy but rather mixes different specialization relations (e. g. is-a, part-of, contained-in) Such resources often very useful as starting point Example: Open Directory hierarchy, containing more then 400, 000 hierarchically organized categories and available in RDF format Chapter 7 A Semantic Web Primer

Linguistic Resources l l Some resources were originally built not as abstractions of a particular domain, but rather as linguistic resources These have been shown to be useful as starting places for ontology development – 21 E. g. Word. Net, with over 90, 000 word senses Chapter 7 A Semantic Web Primer

Ontology Libraries l Attempts are currently underway to construct online libraries of online ontologies – – 22 Rarely existing ontologies can be reused without changes Existing concepts and properties must be refined using rdfs: sub. Class. Of and rdfs: sub. Property. Of Alternative names must be introduced which are better suited to the particular domain using owl: equivalent. Class and owl: equivalent. Property We can exploit the fact that RDF and OWL allow private refinements of classes defined in other ontologies Chapter 7 A Semantic Web Primer

Lecture Outline 1. 2. 3. 4. 5. 6. 23 Introduction Constructing Ontologies Manually Reusing Existing Ontologies Semiautomatic Ontology Acquisition Ontology Mapping On-To-Knowledge SW Architecture Chapter 7 A Semantic Web Primer

The Knowledge Acquisition Bottleneck l l Manual ontology acquisition remains a timeconsuming, expensive, highly skilled, and sometimes cumbersome task Machine Learning techniques may be used to alleviate – – 24 knowledge acquisition or extraction knowledge revision or maintenance Chapter 7 A Semantic Web Primer

Tasks Supported by Machine Learning l l l 25 Extraction of ontologies from existing data on the Web Extraction of relational data and metadata from existing data on the Web Merging and mapping ontologies by analyzing extensions of concepts Maintaining ontologies by analyzing instance data Improving SW applications by observing users Chapter 7 A Semantic Web Primer

Useful Machine Learning Techniques for Ontology Engineering l l l 26 Clustering Incremental ontology updates Support for the knowledge engineer Improving large natural language ontologies Pure (domain) ontology learning Chapter 7 A Semantic Web Primer

Machine Learning Techniques for Natural Language Ontologies l Natural language ontologies (NLOs) contain lexical relations between language concepts – l The state of the art in NLO learning looks quite optimistic: – – 27 They are large in size and do not require frequent updates A stable general-purpose NLO exist Techniques for automatically or semi-automatically constructing and enriching domain-specific NLOs exist Chapter 7 A Semantic Web Primer

Machine Learning Techniques for Domain Ontologies l l l They provide detailed descriptions Usually they are constructed manually The acquisition of the domain ontologies is still guided by a human knowledge engineer – – 28 Automated learning techniques play a minor role in knowledge acquisition They have to find statistically valid dependencies in the domain texts and suggest them to the knowledge engineer Chapter 7 A Semantic Web Primer

Machine Learning Techniques for Ontology Instances l l l Ontology instances can be generated automatically and frequently updated while the ontology remains unchanged Fits nicely into a machine learning framework Successful ML applications – – – 29 Are strictly dependent on the domain ontology, or Populate the markup without relating to any domain theory General-purpose techniques not yet available Chapter 7 A Semantic Web Primer

Different Uses of Ontology Learning l Ontology acquisition tasks in knowledge engineering – – – l Ontology maintenance tasks – – – 30 Ontology creation from scratch by the knowledge engineer Ontology schema extraction from Web documents Extraction of ontology instances from Web documents Ontology integration and navigation Updating some parts of an ontology Ontology enrichment or tuning Chapter 7 A Semantic Web Primer

Ontology Acquisition Tasks l Ontology creation from scratch by the knowledge engineer – l Ontology schema extraction from Web documents – 31 ML assists the knowledge engineer by suggesting the most important relations in the field or checking and verifying the constructed knowledge bases ML takes the data and meta-knowledge (like a metaontology) as input and generate the ready-to-use ontology as output with the possible help of the knowledge engineer Chapter 7 A Semantic Web Primer

Ontology Acquisition Tasks(2) l Extraction of ontology instances from Web documents – – 32 This task extracts the instances of the ontology presented in the Web documents and populates given ontology schemas This task is similar to information extraction and page annotation, and can apply the techniques developed in these areas Chapter 7 A Semantic Web Primer

Ontology Maintenance Tasks l Ontology integration and navigation – l l Updating some parts of an ontology that are designed to be updated Ontology enrichment or tuning – 33 Deals with reconstructing and navigating in large and possibly machine-learned knowledge bases This does not change major concepts and structures but makes an ontology more precise Chapter 7 A Semantic Web Primer

Potentially Applicable Machine Learning Algorithms l l Propositional rule learning algorithms Bayesian learning – l l First-order logic rules learning Clustering algorithms – 34 generates probabilistic attribute-value rules They group the instances together based on the similarity or distance measures between a pair of instances defined in terms of their attribute values Chapter 7 A Semantic Web Primer

Lecture Outline 1. 2. 3. 4. 5. 6. 35 Introduction Constructing Ontologies Manually Reusing Existing Ontologies Semiautomatic Ontology Acquisition Ontology Mapping On-To-Knowledge SW Architecture Chapter 7 A Semantic Web Primer

Ontology Mapping l l l 36 A single ontology will rarely fulfill the needs of a particular application; multiple ontologies will have to be combined This raises the problem of ontology integration (also called ontology alignment or ontology mapping) Current approaches deploy a whole host of different methods; we distinguish linguistic, statistical, structural and logical methods Chapter 7 A Semantic Web Primer

Linguistic methods l l 37 The most basic methods try to exploit the linguistic labels attached to the concepts in source and target ontology in order to discover potential matches This can be as simple as basic stemming techniques or calculating Hamming distances, or it can use specialized domain knowledge (e. g. the difference between Diabetes Melitus type I and Diabetes Melitus type II is not a negligible difference to be removed by a small Hamming distance) Chapter 7 A Semantic Web Primer

Statistical Methods l l l 38 Some methods use instance data, to determine correspondences between concepts A significant statistical correlation between the instances of a source concept and a target concept, gives us reason to believe that these concepts are strongly related These approaches rely on the availability of a sufficiently large corpus of instances that are classified in both the source and the target ontologies Chapter 7 A Semantic Web Primer

Structural Methods l Since ontologies have internal structure, it makes sense to exploit the graph structure of the source and the target ontologies and try to determine similarities, often in coordination with other methods − 39 If a source target and a target concept have similar linguistic labels, then the dissimilarity of their graph neighborhoods could be used to detect homonym problems where purely linguistic methods would falsely declare a potential mapping Chapter 7 A Semantic Web Primer

Logical Methods l l 40 The most specific to mapping ontologies A serious limitation of this approach is that many practical ontologies are semantically rather lightweight and thus don’t carry much logical formalism with them Chapter 7 A Semantic Web Primer

Ontology-Mapping Techniques Conclusion l l 41 Although there is much potential, and indeed need, for these techniques to be deployed for Semantic Web engineering, this is far from a well-understood area No off-the-shelf techniques are currently available, and it is not clear that this is likely to change in the near future Chapter 7 A Semantic Web Primer

Lecture Outline 1. 2. 3. 4. 5. 6. 42 Introduction Constructing Ontologies Manually Reusing Existing Ontologies Semiautomatic Ontology Acquisition Ontology Mapping On-To-Knowledge SW Architecture Chapter 7 A Semantic Web Primer

On-To-Knowledge Architecture l Building the Semantic Web involves using – – – l 43 the new languages described in this course a rather different style of engineering a rather different approach to application integration We describe how a number of Semantic Web-related tools can be integrated in a single lightweight architecture using Semantic Web standards to achieve interoperability between tools Chapter 7 A Semantic Web Primer

Knowledge Acquisition l Initially, tools must exist that use surface analysis techniques to obtain content from documents – Unstructured natural language documents: statistical techniques and shallow natural language technology – Structured and semi-structured documents: wrappers induction, pattern recognition 44 Chapter 7 A Semantic Web Primer

Knowledge Storage l l The output of the analysis tools is sets of concepts, organized in a shallow concept hierarchy with at best very few cross-taxonomical relationships RDF/RDF Schema are sufficiently expressive to represent the extracted info – – 45 Store the knowledge produced by the extraction tools Retrieve this knowledge, preferably using a structured query language (e. g. RQL) Chapter 7 A Semantic Web Primer

Knowledge Maintenance and Use l A practical Semantic Web repository must provide functionality for managing and maintaining the ontology: – – – l There must be support for both – – 46 change management access and ownership rights transaction management Lightweight ontologies that are automatically generated from unstructured and semi-structured data Human engineering of much more knowledge-intensive ontologies Chapter 7 A Semantic Web Primer

Knowledge Maintenance and Use (2) l Sophisticated editing environments must be able to – – – l The ontologies and data in the repository are to be used by applications that serve an end-user – 47 Retrieve ontologies from the repository Allow a knowledge engineer to manipulate it Place it back in the repository We have already described a number of such applications Chapter 7 A Semantic Web Primer

Technical Interoperability l l 48 Syntactic interoperability was achieved because all components communicated in RDF Semantic interoperability was achieved because all semantics was expressed using RDF Schema Physical interoperability was achieved because All communications between components were established using simple HTTP connections Chapter 7 A Semantic Web Primer

On-To-Knowledge System Architecture 49 Chapter 7 A Semantic Web Primer