Change - Ontology Development

Created on Jan. 31, 2013, 12:23 a.m. by Hevok & updated on March 13, 2013, 10:08 a.m. by Hevok

Ontology development is the development of sets of terms (words or phrases) that facilitate searching, categorizing, and visualizing Content. ¶
¶
In order to create an unifying conceptional framework for Aging Research, we are constructing a new concept of the Semantic Web. We are building on the top of other domains, especially those in regard to biology and engineering. For example we are learning from the numerous biomedical ontologies as well as ISO 15926 as they did a lot of work regarding Ontologies. ¶
¶
Three steps are necessary: ¶
¶
1. Treat a living object as an engineering object instead of a living one. Which means each Ontology should be designed for a particular semantic processing. ¶
2. Create an Ontology that allows to combine smaller knowledge-bases, in other words, to have some sort of meta-ontology. ¶
3. Define many models and the ways how the combine these models (like AtlanMod group is doing). ¶
¶
We need to move from low level semantic triples (e.g. RDF elements) to semantic Templates and Patterns. Template is a some sort of a primitive agent - has own internal logic, it consumes input data and does some action (data transformation or initiation of another action), while Pattern produces new knowledge - a standard complex semantic query. ¶
¶
The first point means the work should be case-based. The second and the third points mean that while it is case-based, and if a case is on the system level, we also should think about at a super-system level - how these case-based systems can be combined with each other. ¶
¶
1. We are treating organisms as machines which can be reverse-engineered, then we are doing SENS. ¶
2. We are building an Ontology to classify all the resources and Datasets on Aging as well as all the necessary web technologies. ¶
3. We will try many different models and apply natural selection, but we will also enable to convert any good format into another via a unification concept. We will also take advantage of the well established Semantic Web Standards and provide conversions into them and vice versa. ¶
¶
In our vision Ontology Engineering shall be as easy as just witting English text with some standardization and become even as simple as a visual attractive computer game, so that even kids can grasp the Concepts quickly.The Ontology Development Oriented Activities consists of three different steps. Firstly the Pre-Development activities, basically everything that is done before the Ontology itself is developed. Then the development activities and thirdly what is done after development, the post-development Activities. ¶
¶
For the Pre-Development Activities on has first to do an Environment study. One has to take a look for what Software Platform is the Ontology designated for. So which Application should use and apply the Ontology. It is an environment study because the effort that one invest into the development of the Ontology of course depends upon the Environment. ¶
¶
Then one has to make sure that everything is feasible with a Feasibility study. Can an Ontology really be developed with the means that are available. Does it make sense to develop the Ontology? Does it makes sense in terms of is it cost-efficient. Can we really afford to make this Ontology or is there some other solution that can avoid developing this Ontology. ¶
¶
So the environment and Feasibility Study must be clear before the development starts. ¶
¶
In the Development Stage one starts with the Specification. There one asks for example why is the Ontology developed, what is the benefit and who are the end users for which the Ontology and the Application around the Ontology is intended for. When one has asked these question one starts with a Conceptualizations where one tries and starts to build up a Model, which means one structures the Domain Knowledge into a Conceptual Model. For these one need the help of Domain Experts, because the Knowledge Engineers might not have the Knowledge of the Domain that one wants to represent with the Ontology. When one has the Conceptualization one tries to formalize this. First one does it with a semi-computable Model in First Order Logic, but for the Implementation in the end one needs a computable Model in some Ontology Representation Language. ¶
¶
Then the Development does not stop as one enters the Post-Development Phase. ¶
¶
In the Post-Development Stage one has to think about Maintenance for example. Of course one need maintenance as the World that one has modeled with the Ontology does not stop and changes. It actually Changes continuously. Therefore the Ontology has to be adapted. Thus one has to consider maintenance Routines. When for example are updates and Adjustments of the Ontology necessary. ¶
¶
Also one has to consider Use and Reuse, so which are the designated Applications that use the Ontology and are there also unplanned Applications that can also make use of the Ontology. During the Life-Cycle when the Ontology is already developed, then in the end there might up new Applications that might take use of it. Therefore Reuse belongs to the post-development Process. ¶
¶
1. Pre-Development ¶
2. Development ¶
3. Post-Development ¶
¶
1. Pre-Development ¶
- Environment Study ¶
+ What is the designated Software Platform for the Ontology? ¶
+ Which Applications should use the Ontology? ¶
- Feasibility Study ¶
+ Can the Ontology really be developed? ¶
+ Does it makes sense to develop the Ontology? ¶
2. Development ¶
- Specification ¶
+ Why is the Ontology developed, what is the benefit and who are the end-users? ¶
- Conceptualization ¶
+ Structuring Domain Knowledge in a conceptual Model ¶
- Formalization ¶
+ Formalize conceptual Model in (semi-)computable Model ¶
- Implementation ¶
+ Construction of a computable Model in an Ontology representation Language ¶
3. Post-Development** ¶
- Maintenance ¶
+ Update and adjustment of the Ontology (if necessary) ¶
- Use/Reuse ¶
- Usage of the Ontology within the designated Applications/Ontologies ¶
¶
¶
¶
¶