With the proliferation of information systems used to collect, store, process, and distribute information, most modern organizations have too much rather than too little information – what they lack is the time and experienced staff to effectively use information already available.

An organization’s best experts develop unique capabilities for reasoning about particular subject areas. Applying this knowledge enables them to skillfully use information to make effective decisions.

An expert’s knowledge often depends on personal insight and internalized experience, making it difficult to extend its use beyond the individual’s reach and availability.

Knowledge-Based Systems

Knowledge engineering is the process of eliciting knowledge from experts and integrating it into an explicit representation that a computer system can understand and apply.

Developing software applications that apply expert knowledge solve various problems within the organization is very challenging. Such systems require explicit representation of expert knowledge, based on which they have to be able to handle unstructured information, and unusual situations.

Knowledge Engineering Services -- A Core Capability for ASI

ASI has almost 20 years of experience building knowledge-based systems.

Knowledge Engineering Process

At a high level, this is a two step process:

Knowledge Elicitation -- transform raw knowledge into organized knowledge.

Knowledge Implementation -- transform organized knowledge into operational knowledge.

Knowledge Elicitation

Knowledge elicitation is the process of capturing knowledge through a series of steps, procedures, subject matter expert interviews and other related activities that provide information that can be used to understand the domain area and its relationships to the overall project goal.

Natural Knowledge - Knowledge in its raw form, such as an expert’s accumulated experience and ability to apply it when reasoning about a domain (whether consciously or unconsciously).

Organized Knowledge - Knowledge organized so that it can be easily understood by humans.

Knowledge Elicitation Activities include:

The knowledge engineer works closely with the domain expert to capture the expert's knowledge in a knowledge base.

Iterative, collaborative process

Identify goals and scope

Capture and organize knowledge from interviews, research

Identify data sources & analysis algorithms

Define system actions & user interaction

Access to subject matter experts is critical.

Access to experts is required throughout the engineering lifecycle.

Our knowledge engineers work very closely with your subject matter experts to elicit and capture organized knowledge.

Knowledge Implementation

How do I know it will work? How will you test the system? The Pre-Act® plan-goal graph, concept graph, and situated script representations are all designed to capture the assessment, planning, and execution processes that a human carries out; in fact, part of their purpose is to allow a human to understand the system’s behaviors. These graph structures and their explicit representations of constraints, relationships, and alternatives lend themselves readily to evaluation and verification by domain experts. This evaluation would be followed by a test of the prototype using simulated data, if possible, and field tests of the solution prototype. Even during field testing, the domain experts can continue to monitor the system to understand the reasons for each agent’s actions.

Integrated Knowledge Environment (IKE) - PreAct now comes with IKE, our GUI-based Integrated Knowledge Environment, to cut your greatest cost for building intelligent systems: the knowledge engineering. Good tools make knowledge engineering faster and cheaper. ASI focuses on producing highly usable tools for building knowledge bases for PreAct, such as by providing a palette of knowledge constructs for designing domain-specific planning capabilities.