An antecedent “comes before”. In NetWeaver an antecedent is a node that supplies data or some evaluation up to a higher level node (a dependent node).
An argument is an expression that is used to evaluate the value of a data link. Arguments come in two flavors: crisp and fuzzy.
An argument list is collection of arguments. Some proxies and all data links use argument lists. All arguments are owned by a data link. Arguments in a proxy's argument list are owned by its data link, so the same argument can be efficiently used in multiple proxies. This also means any change made in an argument is global: the change affects every place the argument is used.
synonym for topic
A child node is a node that is attached directly to and beneath the subject node (its parent node). The child node “hangs” from its parent node. In this relationship there is always only one parent and can be multiple children, if supported by the parent node.
A self-referential answer to a question. See circular reasoning.
NetWeaver does not allow connections that would result in circular reasoning.
A data link is a topic type for bringing data into the knowledge base. Data links come in two flavors: simple and calculated. Simple data links bring elemental data directly into the knowledge base. Calculated data links process data through an equation represented by a network of nodes. Either flavor of data link can be compared to an argument from its argument list to calculate a trueness level or (if allowed in the situation) used directly in calculations.
A dependency network is a formal logical representation of how system states at one level of a conceptual model are affected by, or dependent on, other antecedent states. Also known as a “goal”. more...
A dependent is a node that receives data or some evaluation from another node (an antecedent node).
synonym for subject matter expert
With NetWeaver2 comes a new documentation system:
Fuzzy logic is an analysis or control methodology that determines degree of set membership (trueness) rather than “is” or “is not” (true or false). Think “shades of gray” rather than “black and white”. more...
Fuzzy logic is used extensively through NetWeaver.
synonym for dependency network
The hint is a brief description of the topic. In the case of data links it should be the question to be asked to elicit the data. In earlier versions (pre 2) of NetWeaver the “comment” documentation field filled this role.
A hyperlink is a connection to external information. It is represented as a typical URL and can refer to a web-based or file-based document. When a hyperlink button is pressed or an internal jump to hyperlink command is run, NetWeaver attempts to launch your default browser to the hyperlink.
In NetWeaver influence is short hand for “influence of missing data”. NetWeaver has built in capability to calculate the relative influence missing data has on outcomes. This is extremely important for prioritizing efficient and rational querying and data collection. more...
Knowledge engineering is the process of eliciting knowledge, representing the knowledge in some codified way, and verifying the validity of that knowledge representation.
A network is a system of connected nodes such as a dependency network or the equation in a calculated data link. A network is more than representative. It is also a conduit for messages. NetWeaver uses networks to communicate things like data needs and evaluation messages between nodes.
A node is any NetWeaver object that can be connected to another NetWeaver object.
A parent node is the node that is above and directly attached to the subject node. The nodes that “hang” just below a parent node are called its child nodes.
A problem domain is a virtual space within which all pertinent data and fuzzy membership functions are contained that describe our current understanding of the problem domain.
A proxy is a representative or a “stand-in” for something else. In NetWeaver proxies are nodes that represent a topic node in a network.
The following nodes are proxy nodes:
A subject matter expert is someone who has a proficient understanding of the issues of the problem or knowledge (the problem domain) that is being captured into a knowledge base.
A topic is a high level node. Topics have also been referred to as “boss” nodes due to their relationship to proxy nodes. Each topic has a unique proper name. Topics can be any of the following types:
Weight is an numeric attribute of all nodes, however it is only used in certain special situations (see the documentation for individual nodes). If a node's weight is not explicitly assigned it defaults to 1.