Ontological Engineering
Time: Mondays, 4-6:50pm, Fall 2014
Room: 322 Clemens, UB North Campus
Department of Industrial and Systems Engineering: IE 500 (Section 001). Registration number 12656
Cross-listed with:
- Department of Computer Science and Engineering: CSE 510. Registration number 23684
- Department of Philosophy: PHI 598. Registration number 22690
Instructors: Barry Smith and Ron Rudnicki
Office hours: By appointment via email at [1] and [2]
The Course
This is, as far as we know, the first ever course on Ontological Engineering to be offered in a US university. It was first taught in 2013, and videos, presentations and reading materials from the 2013 class are available here: Ontological Engineering 2013. The course provides an introduction to the methods and uses of ontological engineering, focusing on applications in the areas of military intelligence, healthcare, and finance. It will provide an overview of how ontologies are created and used, together with practical experience in the development of OWL ontologies and in the use of associated web technology standards. It will also address some of the human factors underlying the success and failure of ontology projects, including issues of ontology governance and dissemination.
The course will be built out of 3-hour sessions, each of which will involve 2 hours of lecturing and discussion and 1 hour of practical experience with ontology editing software and other Semantic Web technologies.
The course will feature occasional guest lectures by leading ontologists from Buffalo and elsewhere.
Background
Ontologies are an important tool in all areas where data is collected and described by different groups in different ways. Ontologies provide taxonomy-based computerized lexica used to describe diverse bodies of data. They thereby help to aggregate and compare data, to make data more easily discoverable, and to allow large bodies of data to be more effectively searched and analyzed. Ontologies also play an important role in the so-called Semantic Web, where the Web Ontology Language (OWL) forms a central building block in the stack of web technology standards created by the World Wide Web Consortium (W3C).
UB ontologists are involved in a variety of national and international projects in the military, healthcare, bioscience, transport and financial domains. There is an acknowledged shortage of persons with ontological engineering expertise in all these fields, and in related fields such as journalism, manufacturing and government administration.
August 25: Introduction to Ontology
What is an ontology?
How are ontologies used?
What are the differences and interrelations between ontology (philosophy), ontology (science), and ontology (engineering)?
Background materials from last year's class:
September 8: Big Data and How to Overcome the Problems it Causes
We are living in a world of big data. To find our way around this world, we need to identify and integrate the data that is important to our needs. The problem is that data is collected always from different perspectives, with different levels of detail, different granularities for example of space and time, and different communities use different technologies and different terminologies when collecting their data. This session provides an introduction to the problems of data fusion. Strategies to address these problems:
- linked open data
- mashups
- crowdsourcing
- data fusion
We outline some of the successes and failures of these different strategies, and introduce some of the features peculiar to the ontological approach underlying much of the work on data fusion taking place in UB, as a preparation for later sessions in this class.
September 15 The Semantic Web
Web 2.0: The Vision
The Web Ontology Language
Linked Open Data
Universal Resource Identifiers (URIs)
Web 2.0: The Reality
The term "Semantic Web" was introduced by Tim Berners-Lee and others in the late 1990's (1, 2) and first popularized in a paper in 2001 in Scientific American (see below). Berners-Lee summarizes the idea as "a web of data that can be processed directly and indirectly by machines", an extension of the web of documents primarily intended for consumption by people.
- Semantic Web Vision and History Slides Video
- Technology of the Semantic Web Slides Video
- Applications Slides
- Tim Berners Lee Scientific American Articles :
September 22: Use of Ontologies in Tracking Systems
Presenter: Werner Ceusters
A referent tracking system (RTS) is a special kind of digital information system that is designed to keep track of both (1) what is the case in reality and (2) what is expressed in other information systems about what is believed to be the case in reality. An RTS also keeps track of how changes in the information system correspond to changes in the reality outside that system. We will provide an introduction to referent tracking and its implementations. Reading: How to track absolutely everything?
Background
- Basics of Referent Tracking (RT) Video
- RT and Video Surveillance Video
- RT and Data descriptions Video
September 29: How to Build an Ontology
Some examples of simple ontology building
An overview of ontology research in Buffalo
- Military ontology
- Geospatial Information Systems
- Genomics
- Electronic Health Records
- Demographics
Some examples of how ontologies are used
Background
- LTC William Mandrick: Roots of Ontology in the Military Slides Video
- LTC Mandrick: Operational Ontology: The Future of the Military Decision Making Process (MDMP) Slides Video
October 6: Basic Formal Ontology
Why a standard ontology architecture is needed
An introduction to Basic Formal Ontology (BFO)
BFO and its competitors
Building ontologies with BFO
Background
October 13: The Information Artifact Ontology (IAO)
Information Artifacts: Publications, databases, passports, emails
The Email Ontology
The FRBR Library Ontology
The Dublin Core
IAO-Intel
October 20: Ontologies for Image and Sensor Data
October 27: The Science of Document Informatics
What is a document?
What can we do with documents?
What can we do with digital documents that we can't do with paper documents?
What is a diagram?
How can we extend the technology of optical character recognition (OCR) to comprehend also the graphical content of documents?
Background:
November 3
November 10
November 17
November 24: Presentations of Student Projects 1
December 1: : Presentations of Student Projects 2
The materials provided below derive from Ontological Engineering 2013
Ontology and Information Engineering in the Healthcare Domains
- Health care today rests increasingly on the proper use of data deriving from different sources (data pertaining to genes, diseases, symptoms, drugs, medical devices, procedures, hospital infections and other adverse events, hospital management, billing, reporting, and many more). We provide an introduction to the world of healthcare data management, with special emphasis on the role of ontologies and standard terminologies.
- Informatics and Obamacare Slides Video
- Electronic Diseases Slides
- Healthcare Information Management Slides
- Strategies for Data Integration Slides
September 30: Ontology and the Semantic Web
October 7: Military Ontology
October 14: An Overview of Ontology Projects in Financial Domains
- 21. Barry Smith: An Introduction to BFO's Treatment of Information Artifacts Slides Video
- 22. Charles Hoffman: An Overview of Ontology Projects in Financial Domains and Digital Financial Reporting Slides Video
- 23. Charles Hoffman: Financial Report Ontology: The Big Picture Video
Background reading:
Background slides on FIBO, the Financial Industry Business Ontology
Background slides on FRO, the Financial Report Ontology
October 21: Ontologies in Manufacturing: Pitfalls and Promise
- 23. Leonard F Jacuzzo: Semantically Enhanced Manufacturing Analytics Slides Video
- 24. Kumar Madurai: Linked Data in the Enterprise Slides Video
Ontology shows promise in the manufacturing domain. Foundational ontologies such as BFO allow for robust modeling of an entire product life-cycle, thereby enhancing knowledge management, product development, and process refinement. Automated manufacturing requires data describing each instance of a manufacturing process. Used correctly, this data facilitates predictive analytics and root cause analysis. Process and product ontologies focus analysis helping to avoid spurious correlations. Though Semantic Technology allows for computation utilizing ontologies, the embryonic state of this technology often requires sacrificing ontological rigor to achieve real-time data usage. This two-part lecture explores the promise of ontology in manufacturing and strategies for avoiding pitfalls one can face.
Preliminary Readings on Manufacturing Ontology
October 28: Optimization and Fusion
- 25. Moises Sudit: Ontology and Human Intelligences in Optimization and Fusion. Parts 1 and 2
- 26. Barry Smith: BFO and the Command Post of the Future
- 27. Moises Sudit: Ontology and Human Intelligences in Optimization and Fusion. Parts 3 and 4
November 4: Ontology and Natural Language Processing
- 28. Jillian Chavez: A Survey of Natural Language Processing (NLP) Slides
Jillian Chaves has been a computational linguist/language engineer with CUBRC, Inc., since 2012. She holds a Master’s Degree in Linguistics from the University at Buffalo.
November 11: Ontology and Information Fusion Research
- 29. James Llinas: Introduction and Overview of Data and Information Fusion Slides
November 18: The Role of Ontologies in Taming Big Data
- 30. Tanya Malyuta (CUNY): Ontologies vs. Data Models Slides Video
- 31. Tanya Malyuta (CUNY): Horizontal Integration of Intelligence Data Slides Video
Tatiana Malyuta, PhD, is Principal Data Architect and Researcher at Data Tactics Corporation and an Associate Professor of the New York College of Technology of CUNY. She is a subject matter expert in data design and data integration. Recently she has been working on integrated data stores on the Cloud within the context of the Army's Distributed Common Ground System (DCGS-A).
Presentations of Student Projects from 2013
- Jordan Feenstra and Yonatan Schreiber: Music Ontology
- Yi Yang and Jeon-Young Kang: GIS Ontology
- David Lominac: Customer Ontology
- Lucas Mesmer: Manufacturing Ontology
- Travis Allen: Twitter Ontology
- Chad Stahl: Chemical Manufacturing Ontology
- Brian Donohue and Neil Otte: Personality Ontology
- Kevin Cui: GIS Data Model Ontology
- Xinnan Peng: Manufacturing Ontology
- John Beverley: Thermodynamic Equilibrium Ontology
- Paul Poenicke: Gettier Problem Ontology
- Adam Houser: Game Artifact Ontology
- William Hughes and Michael Moskal: Unmanned Aerial Vehicle Ontology
- Kanchan Karadkar: Supply Chain Management Ontology
- Norman Sung: Musical Genre Ontology
Guidance for Presentations and Reports
- Examples of what to include
- Statement of scope of the ontology
- The true path rule
- Identification of existing ontologies
- Explanation of how your ontology differs from (or incorporates) these
- Screenshots of parts of the ontology with some examples of important terms and definitions
- Summaries of potential applications of the ontology
- Evaluation
- Completeness
Grading and Related Policies and Services
All students will be required to take an active part in class discussions throughout the semester. In addition they will be required to design and complete an ontology project, including written description, and brief presentation of the project in class. Students enrolled in the practical segment will be required to create a Protégé file to accompany their ontology project, and also to complete quizzes designed to gauge developing competence in the use of the Protégé Ontology Editor and SPARQL query language.
For 3 credit hour students, your grade will be determined in five equal portions deriving from:
- 1. class participation (1.5% per class attended),
- 2. results of two quizzes relating to the lab portion of the course
- 3. written description of ontology project (3000 words; deadline December 2),
- 4. Protégé ontology file (deadline November 25),
- 5. class presentation.
For 2 credit hour students, your grade is determined as follows:
- 1. class participation (1.5% per class attended),
- 2. written description of ontology project (4000 words; deadline December 2) (50%),
- 3. class presentation (30%).
For policy regarding incompletes see here
For academic integrity policy see here
For accessibility services see here
