Decision Mtg 12 Agenda
From Decision XG
Decision Meeting #12 Goal & Agenda
The dial-in information for this week's meeting is here. The Decision Incubator Activity is intended to explore the use cases and design of a Decision Representation format (e.g. in XML, RDF, OWL…) available on the wiki along with its use by participants in testbed settings for experiments and exercises. Each meeting will be an opportunity to summarize progress & contributions to the wiki, review next steps and action items, and to review the current representation format.
Summary from the Last Meeting
At the last meeting (see minutes), we discussed four major topics: (1) progress on developing the core decision model; (2) implied patterns of decision components; (3) the status of implementing a tool for Leveraging Open Linked Data in an Interoperability Testbed; and (4) current progress with describing our incubator work in a paper. Discussion began with the 4th agenda topic, which everyone is busy contributing to, the paper for the Workshop on Ontology Design Patterns (WOP 2010), see http://ontologydesignpatterns.org/wiki/WOP:2010. Eva recommended a position paper that would describe the vision, why the decision format is useful and how it could be used in practice. Jeff inquired about the focus being more technical or tutorial/motivational and the consensus was to steer it toward the latter. Eva graciously agreed to present the paper at the workshop, if it is accepted. Marion has provided many helpful portions of content and suggestions. Jeff agreed to work on contributions to the section on usage scenario.
The discussion then shifted to topic 1, Eva's work on the "core" decision model. Eva has provided Decision.owl for her core model and DecisionCriteria.owl for her extended model. There is also an example available here which contains criteria and weight, which utilize the parameter pattern. The "core" model is the minimum necessary to represent a decision, including using the situation pattern to represent a decision. In this sense, a decision is a state of affairs. At its heart, a decision is a question with some selected options. The options and question are modelled as information objects using the information realization pattern, which allows one to separate the concept from the realization. For example a literary work is a concept which is separate from a particular book which is the realization. What we want to say about the decision as an abstract concept versus the realization will be best determined by modeling real decisions. On example of the distinction might be the decision to purchase a computer which may be repeated several times by the same person or project, so the decision as a concept is distinct from any particular realization. Eva commented that the patterns that Jeff wrote about on the wiki fit well the current "core" decision model. The "Statement with Variables" to represent a question could be considered the realization of the question. So the core decision model could be extended. Marion is working separately on ontologies for cognition and she described the overlap with our efforts. The members agreed that we should start to explore using these models to see how well they work for real decisions.
The discussion shifted to Topic 3, so Jeff discussed his colleagues' progress developing a web-based application to capture decisions in this format. This involves the basic gui, the use of keywords to identify the applicable open linked data sets, the use of the "entries" from the open linked data sets as options, the use of the entry properties as selectable metrics, the application of greater than, less than filters, and the use of weights on those metrics
Goal for this Meeting
At this twelfth meeting of the Decision Incubator, we should explore an iterative process for evolving the current decision models. This should involve taking the designs, instrumenting current tools such as the one Jeff mentioned for making decisions about open linked data, then exploring the tool for a variety of applications. Once this process is set up utilizing the existing interoperability testbed infrastructure and architecture, decisions can be made in applicable domains. The process should include creating open linked data sets where desired or needed. For example, in the emergency management domain, the conversion of EDXL standards into RDF to create linked data sets is something that should be done to support these efforts and to support improved interoperability of these datasets in general.
The following agenda items will be discussed at the next meeting. The topic, time, summary and discussion leader are provided for each agenda item.
(0) Review of Status, Goals & Agenda Overview (5 min, Jeff)
Use of our core component models and patterns list will be mentioned and suggested goals for next few weeks.
(1) Update on "Core Decision" Model for Decisions (15 min, Eva/Jeff)
Any updates to the core decision model will be discussed along with any questions members have about the core model or the patterns upon which they were based.
(2) Suggested Process for Exploring and Validating Models (15 min, Jeff)
Jeff will suggest a process for exploring the core model and extension components through the use of a tool being developed for exploring decision-making with open linked data. The process will be iterative, utilizing the interoperability testbed and concept, and with the suggestion to utilize the emergency management domain.
(3) Conversion of Emergency Management Data into Linked Data format. (15 min, Jeff)
The emergency management domain involves a myriad of decisions which should be shared for improved situational awareness. Conversion of current data formats into RDF would support the exploration of the current decision models in this domain. Jeff will suggest this initiative and discuss the basic components of the current open linked data RDF format and possible extensions and the impact for utilizing the data for decision-making.
(4) Schedule and Future Papers. (15 minutes, Eva/Marion)
The current paper submission to the WOP conference describing the Decision Incubator work will be discussed. Since the paper is a "position" paper, it will deserve an update as tools are developed, the models are developed, and the process moves forward to validate the models. The possible "next" paper will be discussed in this context.