Ontolog Forum
Ontology Summit 2017 Topic Exploration Main Page
Building Blocks Exploration Exercise
During this year's summit we were presented with many uses of ontologies, machine learning, and artificial intelligence (in their various guises). One observation from all the work presented is that there are many many more possible applications and new capabilities (and of course areas for further research) still to be discovered.
The intent of this exercise is to explore the use of some 'building blocks' to,
- a) Craft a solution to a problem (or class of problems)
- b) Clarify what issues or 'details' still need to be overcome in realizing a solution
- c) Devise a new capability
- d) Something else
The building blocks (listed below) are devoid of specifics and denote only a capability and this list is not exhaustive. Each team can add additional building blocks or refinements of those listed.
The goal is to combine these 'building blocks' (aka capabilities) in novel ways. As analogy think of Lego blocks and how they can be put together to form a variety of objects. Though this can be construed as an exercise in design or engineering, don't be overly focused on (apparently) non-significant details.
Among potential 'problems' for consideration
- 0) Mining the Ontolog Forum e-mail and Summit archives (for knowledge)
- 1) Devising a better way to craft legislation
- 2) Analyzing or developing regulations (from legislation)
- 3) Mining one more journals in a particular domain (for knowledge)
- 4) Mining any large corpus of materials for commonality and the context of such commonality
- 5) Identifying context or sentiment
Building Blocks/Capabilities
Reasoning
- Deductive
- Inductive
- Analogical
- Abductive
Machine Learning
- Supervised
- Unsupervised
- Statistical Relational Learning
Ontologies
- Static/Pre-Defined
- Dynamic
Neural Nets
Genetic Algorithms
Natural Language Process (NLP)
- Latent Semantic Analysis
- Syntactic Analysis – Part-of-Speech (POS) tagging,
- Sentence Breaking/sentence boundary disambiguation, etc.
- Named Entity Recognition
- Relationship Extraction
- Sentiment Analysis
Speech/Audio Process
- Speech to Text
- Text to Speech
- Speech/Audio to visualization
Image Processing
- Image Recognition
- Image Creation
- Image Editing
Data Analytics
- Data Mining
- Statistical Analysis
Resources to Consider
- Existing ontologies - BFO, DOLCE, FIBO, ....
- Sensor Data - real time or persisted
- Existing data sources
- WordNet, VerbNet - currently joined under the title 'SemLink'
- Taxonomies
Ontology Summit 2017 Topic Exploration Group 1
Ontology Summit 2017 Topic Exploration Group 2