Teaching People to Teach Machines

Thematix has gained a reputation as a teacher of note in the area of semantics, natural language processing and ontologies.  Our “Ontology 101” is a hit at trade shows, year after year.  You may enjoy paging through a version of this presentation here:

Introduction to Semantics and Ontology

We can bring our long experience teaching graduate computer science students and industry groups to your firm, to educate a variety of staff in becoming familiar with ontologies and semantic technologies, covering a broad range of topics, including an overview of best practices, tools and techniques used to build semantically enabled systems. For this broad audience, we typically deliver the following syllabus:

1. Introduction
   1. Origins and evolution of semantic technologies
   2. The business case for semantics
   3. Enterprise use cases, applications in science, healthcare and pharma
   4. Emerging, high-profile applications and uses – for question answering, integration, interoperability, data governance, recommendation systems, etc.
2. Ontology Engineering Overview
   1. Ontology basics – definitions, underlying logic fundamentals
   2. Business requirements and use case driven approach
   3. Development methodology including conceptual modeling, terminology work, vocabulary and logic languages, ontology development, provenance, evaluation and maintenance
3. State-of-the-Art
   1. Semantic technologies and machine learning
   2. Current applications survey
      1. Industrial applications
      2. Pharma applications
   3. Significance and implications for Merck
      1. Possible future directions

Ontology Engineering Training

For more immersive course, content is drawn, in part, from an advanced, graduate degree level course (see https://tw.rpi.edu/web/Courses/Ontologies/2016 for an overview of the semester-long course), with roughly 1.5 to 2 days on the ontology engineering methodology, best practices, and hands-on exercises. This is a fairly aggressive syllabus, and will require coordination in advance to ensure that participants have at least a couple of tools loaded on their machines and access to the internet during the sessions.

1. Process
   1. Principles of Business Architecture
   2. Requirements Gathering using Methods from Business Architecture and Use Case Analysis
   3. Terminology Extraction and Analysis
   4. Conceptual Modeling
2. Ontology Development
   1. RDF, RDFS, OWL Language Concepts
   2. Introduction to Formal Logic, with focus on Description Logics
   3. Best Practices in Ontology Development, including Common Patterns, Anti-Patterns
   4. Common Architecture Challenges – Namespaces, Modularization
   5. Using Tools to Check Your Ontology: Syntax, Semantics, and Regression Testing
3. Advanced Ontology Engineering Practices
   1. Provenance, References, and Evidence Collection
   2. Change Management
   3. Ontology Evaluation, Reuse and Extension
   4. Ontology Evolution
4. Question Answering Using RDF, Triple Stores and SPARQL
   1. RDF and Triple Stores
      1. RDF and RDF serializations (XML, TRTL, N3)
      2. Triple Stores, HTTP endpoint
      3. SPARQL
5. SPARQL with reasoning assistance (A-Box queries)
6. OWL 2 Profiles: EL, QL, RL
7. Reasoning as query re-writing
8. SPARQL queries against the T-Box
9. Practical Applications for SPARQL and Triple Stores
   1. Testing Ontologies
   2. Big Data
10. Language Processing
    1. Named Entity Recognition
    2. Information Extraction, Text Analytics
    3. Query Answering, Document Summarization
    4. NLP Technologies
       1. Dictionaries
       2. Vector Spaces
       3. Hidden Markov Models
11. Supervised Learning Engines
12. NLP Pipeline Architectures
13. Morphology
14. Syntax
15. Pragmatics
16. Practical Applications and Resources
17. Advanced NLP Application Overview (dialogue, translation,…)
18. System architecture
19. Data Integration, Migration and Management, including R2RML
20. Development strategies, feasibility assessment
21. Advanced Topics
22. Question answering
23. Decision support and recommendation systems, using reference and operational data, streaming data, and so forth
24. Applying Ontology to NLP
25. Reasoning Beyond Description Logics (g. Flora 2, Answer Sets, Bayesian reasoning, Prolog)