Data processing: artificial intelligence – Knowledge processing system – Knowledge representation and reasoning technique
Reexamination Certificate
1999-08-24
2002-12-10
Black, Thomas (Department: 2121)
Data processing: artificial intelligence
Knowledge processing system
Knowledge representation and reasoning technique
C707S793000, C707S793000, C707S793000, C707S793000
Reexamination Certificate
active
06493697
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates generally to knowledge management systems, and particularly to the development of knowledge base systems.
One environment in which knowledge management systems are particularly useful is the computer product support industry. The computer systems on today's desktops are complicated. They involve many products (hardware and software) from many vendors. The products may or may not operate as expected when configured into a system. In addition, the user guides and references for the products are often incomplete and not always accurate. When end users have problems with their computer systems, they often need help diagnosing and then solving the problem. The computer product support industry has developed in response to that need. When a caller into a technical support organization reports a problem with a product in a certain environment, a technical support representative, sometimes known as an agent, diagnoses and attempts to solve the problem.
However, a mountain of knowledge is necessary in order to provide support for computer products. End users' answers might be found in a public document, or in a customer's or vendor's confidential information, or in a company's bank of general or confidential knowledge. In addition, through support interactions, a company generates a vast array of knowledge, particularly in areas such as product interoperability. Knowledge is always being generated because the resolution to an end user's problem may even need to pieced together from many sources of information, public and private combined.
A computer product support provider's challenge is to handle the increasing technical complexity of support delivery while keeping service quality and customer satisfaction high and costs low. Companies must establish a support infrastructure that enable them to capture, refine, and publish customer service and support information with greater efficiency through a variety of support channels. Adopting a knowledge management approach is an effective means to meet urgent customer demands.
One part of the knowledge management approach is the development and maintenance of knowledge bases as a part of a company's knowledge management system. With the proliferation of information that is needed to run businesses today, many companies are turning to knowledge base systems to store and provide access to its information. Knowledge bases provide a framework for collecting, organizing and refining the full range of information that is both collected and generated daily by a company. Knowledge bases process the new information, transforming it into actionable knowledge, present it in a consistent format, and make it readily available. They make a company increasingly effective in gathering and leveraging “institutional memory.” Thus, knowledge bases provide a company with the opportunity to reuse the knowledge that it collects and creates. Such reuse is beneficial because it allows companies to use its data to conduct is business more quickly and efficiently than previously possible.
While knowledge bases provide some real benefit to companies that invest in their creation, they are expensive in time, resources and money to develop. Many complex issues must be addressed. For example, keeping in mind the work environment in which the knowledge base will operate, developers must decide the subject areas (known as domains) that would benefit from having knowledge about them incorporated into a knowledge base. They must select from which sources should the knowledge be obtained and the extent that an order in which the knowledge base should be seeded prior to activation. In addition, the developers must develop the knowledge base's architecture based on the information that it will hold and the use to which the knowledge will be put. They must develop operational processes for using the knowledge base and integrating it into with the other systems and processes used by the knowledge base user.
It is therefore an object to develop knowledge management systems that allow a company to manage the knowledge it collects and creates, make it available for use in conjunction with the other systems and processes used by the company, and monitor its use. It is a further object of this invention to develop and deploy a knowledge base so that it quickly contains a vast array of information. It is also an object to seamlessly integrate the knowledge base with the other systems and processes used by the knowledge base user, and develop operational processes for keeping the knowledge base updated.
SUMMARY OF THE INVENTION
In accordance with the present invention, there is described a method and a domain suitability indicator for selecting a desired knowledge domain for a knowledge base from a set of potential domains, the set having at least one potential domain. A domain suitability value is developed for each of the potential domains in the set. The value indicates suitability for becoming the desired domain of the potential domain for which the value was developed. The potential domain suitability values are compared and the desired domain selected from the potential domains based on the comparison.
The potential domain suitability values are developed by identifying for the set of the potential domains at least one attribute having an ability to forecast a potential benefit of selecting the potential domain to be the desired domain. For each attribute, an extent of the potential benefit is established for selected instantiations of the attribute, and an attribute valuation system is developed that demonstrates the extent of the potential benefit for each of the selected instantiations. The valuation system is developed by assigning an attribute benefit value to each of the selected instantiations, with each of the attribute values indicative of the extent of potential benefit from selecting the potential domain to be the desired domain.
For each potential domain, an actual attribute value is developed for each attribute of the potential domain by developing an actual instantiation for the domain. The actual instantiation may be identified by identifying a characteristic of the attribute, or, for an attribute that is measurable, the actual instantiation is identified by measuring the attribute. For each attribute of each potential domain, an actual attribute score is assigned, from the attribute valuation system of the attribute, with the actual attribute score based on the actual instantiation. Each of the actual attribute scores are weighted according to goals to develop the actual attribute value, and the actual attribute values for a potential domain are combined to generate the potential domain's potential domain suitability value. In the preferred embodiment, the combining comprises summing the attribute values.
In one embodiment, the attribute may be organized into sub-attributes, and an attribute valuation system developed for the sub-attributes. The sub-attributes could then be treated as attributes when evaluating the potential domains.
Once the desired domain is selected, a more specific desired domain may be selected for the knowledge base by apportioning the desired domain into a plurality of more specific potential domains. The more specific potential domains could be sub-topics of the subject matter of the desired domain, or they could be functionalities of the desired domain. A more specific potential domain suitability value is developed for each of the more specific potential domains in the second set. The more specific potential domain suitability value indicates suitability for becoming the more specific desired domain of the more specific potential domain for which the more specific potential domain suitability value was developed. The more specific potential domain suitability values are compared and the more specific desired domain is selected from the more specific potential domains based on the comparison.
In one embodiment, the method further involves
Haughton Debra Ann
Stier Sharon
Black Thomas
Hirl Joseph P.
Moran Maura K.
Stream International, Inc.
LandOfFree
Method of selecting desired domains and for developing a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of selecting desired domains and for developing a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of selecting desired domains and for developing a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2946606