Knowledge Management Using Orion Making Knowledge Active Knowledge

Knowledge Management Using Orion Making Knowledge Active

Knowledge Is Structure What does that mean? What sort of structure does it imply?

ORION Orion is a knowledge utilisation system based on using an active, undirected structure to represent knowledge Orion covers the full knowledge cycle – creation or capture, marshalling and deployment It is particularly suited to dynamically changing problems, where its ability to modify its own structure comes to the fore

A new approach to knowledge Instead of writing programs, turn the knowledge itself into a computing machine Example: a + b = c Structure Programmatic: if input(a) and input(b) then c=a+b else if input(a) and input(c) then b=c-a else… NYK - Not Yet Known

Use the structure itself Example: a + b = c Values flow through the structure in any direction. The EQUALS operator allows for logical control.

Information Coming the Other Way Example: a + b = c We have a range on one variable, producing a range on another.

The System as a Thinking Tool Orion The user can turn what they know into a shareable and reusable piece of knowledge by describing relations among objects - the system will use the relations any way it needs to, when it needs to. Users can refine their thinking by observing a model of it in operation - does it match reality?

Combining Knowledge Domains Models from different domains can be combined easily because they are undirected each expert operates in their own domain, thinking in ways they are comfortable with. Economy Funding Credit Risk Integrated Model FX

Techniques Knowledge about the problem area is turned into a structure, made up of variables, operators and links. The structure is undirected and extensible, and supports the following problem-solving techniques: • Logical and Numerical Analysis • Ranges of Values • Simulation • Structure Self-modification • Backtracking • Constraint Reasoning • Hypothesising

Ranges of Values Ranges allow you to specify limits in a natural way. The duration is between 4 and 10 days. The cost should be between 1 and 3 million. The expected magnitude lies anywhere from 4. 8 to 6. 3.

Logical & Numerical Analysis Here are logic and numbers interacting together in a plan. If there is insufficient time for Activity_4’s duration, or insufficient budget or resources, it forces itself false and the other alternative becomes True.

Probability Distributions Probability distributions are another important source of technical knowledge. They need not follow simple forms like Normal or Poisson.

Backtracking When there are lots of possibilities to consider, you want the system to try to find a good solution for you. Backtrack allows the system to try something, observe the result, and then undo it and try something else. While trying something, it may need to erect new structure - “build a castle in the air” - then undo all that as well.

Constraint Reasoning Sometimes there may be constraints on a solution, but no analytic way of finding it. Then you have to try different possibilities, while respecting the constraints. If you use the constraints to prune away downstream possibilities, this is Constraint Reasoning. It helps if the structure of the constraints is undirected, so you can start where there are only a few possibilities - where the problem is most heavily constrained.

Using the System

Logic Editor - Used to enter declarative knowledge in textual form. The text is immediately transformed into active model structure.

Network Display The user can: • trace the structure of the network • observe the values in particular components • set and unset values • trace the source of inconsistencies • debug the network by halting propagation and observing states

Analysis and Visualization Drawing Editor Stochastic Editor List Editor

Stochastic Editor The user can • visualize and manipulate distributions and Ndimensional relations. • apply constraints to variables and observe the impact on other variables • construct ad hoc distribution and relation operators

Data Miner One of the shortcomings of existing data mining technologies is that in order to use the findings, the user needs to understand them. The reason is that the technologies used for mining are different to those used in operational systems. With Orion, the same technology is used for both tasks. The Miner actually morphs newly found correlations in the data into an active component of the operational system. The system can start with mined data, then “learn on the run” from new transactions.

Additional tools and facilities • Structural debugger • Script editor • Simulator • Graphing • XML Link • Time Series Analysis module • Structure extraction from text

Applications • Knowledge Management in Organisations • Strategic Planning, Program/Project Management • Financial Analysis, Wealth Management • Risk analysis • Simulation, Workflow, Scheduling • Knowledge Representation

Simulation - Insurance/Reinsurance

Model the Problem Area The model takes the form of the business structure it represents. It is “alive” all the time, every piece of it is visible and accessible. Programmatic approach Orion Household Commercial Property Quota Share treaty Surplus treaty Risk Access treaty Catastrophe treaty Business Domain

Creation and Destruction of Objects When the simulation starts, the portfolio has 10 bond invocations maturing on 2001 to 2010. After a 20 years simulation, where maturing bonds are destroyed and new bonds are purchased if the cash position allows, the portfolio is left with 3 bonds purchased in 2018, 2019 and 2020.

Reinsurance Simulation Household The structure generated for Commercial Property (each column represents a policy) Generated Commercial Property Claims A Surplus treaty Claims paid after the Surplus recoveries Risk Excess treaty Claims paid after the Surplus and Risk. Xs Combine with Household Claims after Quota Share (into Catastrophe treaty) The structure of the reinsurance program is in the form that insurance people understand it – claims flow into a reinsurance treaty, which recovers some of the payments. What remains can be combined and flow into another treaty.

Earthquake Knowledge Model Structure Find distance between site and Epicenter, local conditions, etc. Acceleration attenuation based on magnitude, distance and local site conditions Attenuation Greece Info (GIS) Recorders Intensity/ Damage Events XML Event Relationsbetween acceleration, intensity and damageratio Frequency/ Amplification Relations between magnitude and frequency, building type, number of floors and natural frequency, including comparison and amplification estimates

Analytic Structure - Wave Attenuation A formula for attenuation (note function on left hand side)

Experiential Structure Wood Reinforced Concrete The Sauter curves, linking Intensity, Building Type and Damage Ratio, are loaded into distributions and relations.

Accept Anything Here is KM at work. Need to accept: • Different forms of knowledge - analytic, experiential • Different timescales for validity - yesterday, last month, always • Different sources - journals, consultants, suppliers, customers Integrate it, iron out the inconsistencies, if necessary change it on the run - manage it.

Propagation Experiential submodel Damage Ratio An example of one of many value propagation paths, from event magnitude to damage ratio. Experiential submodel intensity Magnitude – Event submodel Attenuation submodel SAh

Compatible Knowledge Structures A variable can have a PLUS operator on one side and a RELATION on the other, so analysis and experience are easily combined.

Knowledge in Text Much knowledge is held in text. Unless the text can be accurately “read”, that knowledge is unreachable by automation. Techniques such as Text Mining barely scratch the surface, and requiring someone to attempt to pluck out the knowledge into checklists, ontologies or other simple forms is doomed to failure knowledge is much richer than these simplifications allow. Orion can read text and create an Active Structure - an active representation of what the text means.

Active Structure from Text A look at the finish. Date of the Occupy relation in a contract. The date value of the Implied. One is linked both to an “and” objectgroup comprising August 31, 2005, and the date which is the Start. Date of the IS relation of “Tenant’s Work is Substantially complete”. The OBJECTTEST operator created a MIN operator linking the three date values. Note that since the date is not known yet, its value is a range (green), which propagates all the way to the occupy relation.

Orion can handle a wide range of problem areas. Its power comes from its simple form variables, operators, links - and the fact that it is easy to combine structures which do not have a beginning or an end.

How Is It Different? It is based on: • Connection • Activity • Visibility • Extensibility • Self Modification • Undirectedness

It Sounds Too Complexity appears quite quickly. A + B = C is static. X = SUM(List) is already dynamic. “The earthquake struck Sao Paulo last Thursday. ” is just more dynamic. Knowledge in an organisation is dynamic - it is just that humans handle it without a second thought. For a system to handle it well, the system needs some similar properties - activity, connectivity, visibility, selforganisation.

The Benefits Knowledge in many technical areas is hugely dense. ORION brings a logical structure and the ability to represent dense knowledge in a way that is useful to the organisation. Instead of being locked in a few people’s heads, it becomes a shareable resource, giving others the ability to comment on it, criticise it, add to it, link it into an organisational whole.