{"id":56736,"date":"2023-04-19T09:17:36","date_gmt":"2023-04-19T08:17:36","guid":{"rendered":"https:\/\/developer.kore.ai\/v10-1\/?p=56736"},"modified":"2023-04-19T09:17:36","modified_gmt":"2023-04-19T08:17:36","slug":"knowledge-graph-terminology","status":"publish","type":"post","link":"https:\/\/developer.kore.ai\/v10-1\/docs\/bots\/bot-builder-tool\/knowledge-task\/knowledge-graph-terminology\/","title":{"rendered":"Knowledge Graph Terminology"},"content":{"rendered":"

This article explains the terminology for <\/span>building a Knowledge Graph<\/span><\/a> within the XO Platform. This terminology applies to both the Few-Shot and the Ontology KG Models unless otherwise specified. <\/span>Learn more about Knowledge Graph Types<\/span><\/a>.<\/span><\/p>\n

<\/div><\/div>\n

Terms or Nodes<\/span><\/span><\/h1>\n

Terms or Nodes are the building blocks of an ontology and are used to define the fundamental concepts and categories of a Knowledge Graph.<\/span><\/p>\n

You can organize the terms in a hierarchical order to represent the flow of information in your graph. You can create, organize, edit, and delete terms and set term types to facilitate intent identification\u00a0<\/span><\/p>\n

The Knowledge graph is limited to 20 thousand nodes\/terms and 50 thousand FAQs.<\/b><\/p>\n

Node\/Term Hierarchy<\/span><\/span><\/h2>\n

For easier representation, we identify node\/term hierarchy using the following names:<\/span><\/p>\n

Root Term\/Node<\/span><\/h3>\n

The <\/span>Root Term\/Node<\/i><\/b> forms the topmost term of your Ontology. A Knowledge Graph contains only one root node, and all other nodes in the ontology become its child nodes. The <\/span>Root <\/span><\/i>node takes the name of the VA by default, but you can change it as needed. In an Ontology-based KG, this node is not used for path qualification or processing. Path qualification starts from first-level nodes.\u00a0<\/span><\/p>\n

For Ontology-Based graphs, it is best to organize your FAQs into a clear structure, so we don\u2019t recommend adding them at the <\/span>Root <\/span><\/i>node. You can, however, add a maximum of 100 FAQs if required.\u00a0<\/span><\/p>\n

First-level Term\/Node<\/span><\/h3>\n

The immediate next-level nodes after the <\/span>Root <\/span><\/i>node are known as <\/span>First-Level Term\/Node<\/i><\/b>. There can be any number of first-level nodes in a graph. We recommend using first-level nodes to represent high-level terms, such as the names of departments, functionalities, etc. <\/span>For example, in a Travel Assistant, you might have a first-level node called <\/span>Reservation<\/span><\/i>, which can be structured by functionality into subnodes such as: <\/span>Cancel <\/span><\/i>and <\/span>Update<\/span><\/i>.<\/span><\/p>\n

Leaf Term\/Node<\/span><\/h3>\n

Any node at any level starting with the 2nd is called a <\/span>Leaf Term\/Node<\/i><\/b>.<\/span><\/p>\n\n\n\n
Note<\/em><\/strong>: This hierarchical organization of nodes is for your convenience to keep related questions together. The Knowledge Graph Engine does not consider any parent-child relation while evaluating the questions for a match. The hierarchy does not influence the FAQ matching process since all the nodes are considered the same way, irrespective of their position.<\/span><\/em><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\"\"<\/a><\/p>\n

Term\/Node Types<\/span><\/span><\/h2>\n

Functionally, there are three types of Terms\/Nodes.<\/span><\/p>\n