The internet provides a multitude of vendors promising to transform our business. But , we have to know which approaches should be adopt to examine the component technologies that make up a knowledge management system or suite. The analogy of hi-fi is used where each item has a certain function or purpose. The multitude of KM system on offer in the marketplace is seen as a composite variation of a number of these component technologies. Firms may decide to buy different component-of-the-shelf or develop their own tools to meet their needs.
Organising knowledge tools
Ontology and Taxonomy Grubber (1993) defines ontology as “a formal explicit of shared conceptualism”. It helps us on preventing wide variations on understanding or perspective to the same subject. Therefore, we have to developed ‘ontology’ to improve our level of information organization, management, and understanding. In the context of KM tools, the term ontology is often used interchangeability with taxonomy. To clarify the distinction, it’s important to recognize that an ontology is overall conceptualism whereas taxonomy is a scientifically based scheme of classification. An ontology may have non taxonomic conceptual relationship such as ‘has part’ relationship between concepts. In contrast, knowledge taxonomy generate hierarchical classification of terms that are structured to show relationship between them.
When it comes time to implement ontologies and taxonomies there are three options:
· develop the ontology then develop the supporting taxonomy. · develop taxonomy and then develop the over-arching ontology · develop the two in parallel.
Define your scope
The first step in developing the combined ontology and taxonomy is to clearly scope the effort. A clearly defined scope is critical to the success of the effort. The question that can best help shape the scope of the effort is simply, what purpose will the combined ontology and taxonomy serve? The answer to this question serves several purposes: · It sets bounds on the effort. These bounds are necessary to answering the basic managerial questions of how long will it take, and how much will it cost. · It helps identify the primary domains and perspectives to be included. · It should identify the specific business activities that will make direct use of the
ontologies and taxonomies and how the resulting knowledge will be used to accomplish their mission.
Normally the answer to this question will fall into one of three categories: · To serve as a common framework for knowledge sharing. · To enable reuse of existing domain knowledge. · To a better understand what the organization knows by separating domain knowledge from the operational knowledge and making assumptions explicit.
The first two of these, knowledge sharing and reuse are at the heart of most KM initiatives. Answers falling into the third category are indicative of advanced KM or academic initiatives. The following steps are applicable to all three scenarios and are specifically targeted toward those efforts focusing on knowledge sharing and reuse.
Check for Existing Ontologies and Taxonomies
Business operations today are also often dependent on, or required to adhere to one or more industry standards and may interface with applications that make use of existing ontologies or controlled vocabularies. For that reason, it is often best to use preexisting taxonomies and ontologies before launching into an extensive and possibly expensive development effort.
Identify Important Terms
If it is determined that existing ontologies and taxonomies are insufficient to meet the scope of the effort, then it is time to start collecting the raw materials for the new structures. This starts with identifying the key terms that are used to express the knowledge needed to enable specific business activities. At this point, it is important to list of all terms used to make statements or to explain to someone else what is needed to accomplish the business activities included within the scope of the effort. These can normally be found in corporate policy or operational instructions and from people with expertise in the activities. While gathering these terms it is important to stay within the scope of the effort. Incorporating the full lexicon for a large multi-national conglomerate may look like a major accomplishment, but it will not help build an effective knowledge environment if the scope of the effort is smaller. Rather, the broader scope will make it more difficult to marry the concepts of the field to those used by the people you are trying to support.
Define the Class Hierarchy
Armed with the terms and concepts that are critical to expressing the knowledge needed to enable those business activities to be supported, the next step is to define a class hierarchy. Again, there are three ways to do this: · Work from the top down. Start by identifying the general concepts (super-class identification) and then determine which of the others fall within those categories (subclass identification). · Work from the bottom up . Start by developing cluster of related concepts (subclass grouping) and then look for the higher-order concepts that under which a given cluster of concepts might fall (super-class identification) · Work from the top down and bottom up. This starts with the identification of the more important concepts first and then generalize and specialize them appropriately. The authors have found that this approach is the easiest to follow and has a lower risk of getting mired in the “semantic swamp.” When organizing terms and concepts the basic formula is:
If a class A is a super class of class B, then every instance of B is also an instance of A. In other words, the class B represents a concept that is a “kind of” A. As the class structure for your domain begins to take shape, check how well it expresses the way the people that are actually involved in the associated activates actually talk about what they are doing. Make sure that it reflects the actual context in which it will be used. The language of practical ontology or taxonomy should not to require the user to stop and translate between the way they normally think about things and the “official” term. This is also the time to make sure that the vocabulary used to express the hierarchical class structure agrees with the real-world (and agreed upon) vocabulary and accurately serves to classify the information that will need to be exchanged between the people and applications involved in meeting the targeted activities. When these conditions occur, the class hierarchical effectively becomes the upper-layers of the taxonomy. This is a critical factor in insuring on-going semantic alignment between the resulting ontology and taxonomy.