{"id":47701,"date":"2022-10-04T11:50:52","date_gmt":"2022-10-04T10:50:52","guid":{"rendered":"https:\/\/developer.kore.ai\/v10-1\/?p=47701"},"modified":"2022-10-04T18:33:44","modified_gmt":"2022-10-04T17:33:44","slug":"build-a-travel-assistant-knowledge-graph","status":"publish","type":"post","link":"https:\/\/developer.kore.ai\/v10-1\/uncategorised\/build-a-travel-assistant-knowledge-graph\/","title":{"rendered":"Travel VA: Build a Knowledge Graph"},"content":{"rendered":"

This article shows you how to build and understand the working of a Knowledge Graph for your Virtual Assistant. It shows how to convert an FAQ into ontology, add tags, use terms, add alternate questions, synonyms, tags, and traits.<\/span><\/p>\n

For details on the implementation process, please <\/span>refer here<\/span><\/a>.<\/span><\/p>\n

Throughout this use case article, we will be building a Knowledge Graph for a Travel Planning Assistant. This article discusses both the basic setup of a Knowledge Graph, but also the solutions at hand for cases of ambiguity.<\/span><\/p>\n

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

Static FAQ<\/span><\/span><\/h1>\n

Let us start by creating a static FAQ without an ontology.<\/span><\/p>\n

To add a static FAQ, follow the below steps:<\/span><\/p>\n

    \n
  1. Open the Travel Planning Assistant.<\/span><\/li>\n
  2. Navigate to <\/span>Build<\/strong> > <\/span>Conversational Skills > Knowledge Graph<\/strong>.<\/span><\/li>\n
  3. Click <\/span>Create<\/strong> to create a Knowledge Graph.<\/span><\/li>\n
  4. A Knowledge Graph with the name set to the Bot Name is created.<\/span><\/li>\n
  5. Click <\/span>Add Intent<\/strong> to start entering your FAQs.<\/span><\/li>\n
  6. Select <\/span>Intent<\/strong> type as FAQ.<\/span><\/li>\n
  7. Add Question<\/strong> and <\/span>Bot Response<\/strong>; then click <\/span>Save<\/strong>.
    \"\"<\/a>
    <\/span><\/li>\n
  8. Click <\/span>Add Intent<\/strong> on the top-right to add more FAQs.<\/span><\/li>\n
  9. After you enter all the FAQs, click <\/span>Train <\/strong>to train the Knowledge Graph.
    \"\"<\/a><\/span>
    <\/span><\/li>\n
  10. Click <\/span>Talk to Bot<\/strong> at the bottom-right to check the response for the added FAQs.<\/span><\/li>\n<\/ol>\n
    <\/div><\/div>\n

    Tags<\/span><\/span><\/h1>\n

    Tags help filter the FAQs based upon the presence of some key phrases in the user utterance. This helps identify the intent more efficiently and prevents the assistant from providing users with irrelevant responses. We recommend that you add tags to every question, unless you have a large list of FAQs. In this case, you need to create an Ontology (see below for details). To add tags to an FAQ:<\/span><\/p>\n

      \n
    1. Hover over the Question then click the <\/span>edit (\ud83d\udd89)<\/strong> icon.<\/span><\/li>\n
    2. In the <\/span>add tag<\/strong> text box, you can find some suggestions by the Platform based on the phrases present in the question. You can pick one from the list or type your own custom tag.<\/span><\/li>\n
    3. Train<\/strong> the assistant.<\/span><\/li>\n
    4. Talk to the Bot<\/strong>, and use an utterance that points to the FAQ you added tags to. Formulate the request in different ways, to find potential tags to add.\u00a0<\/span><\/li>\n<\/ol>\n

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

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

      Ontology<\/span><\/span><\/h1>\n

      For a better response, you must add tags for every question. But this solution is not feasible in case we have a big list of FAQs or questions with common tags. Ontologies help in this situation because they rely on key words and frequently occurring words\/phrases to organize FAQs in groups and subgroups.\u00a0<\/span><\/p>\n

      Each group can be a node in the ontology. You can have multiple levels of the nodes forming a hierarchical structure. Moving the relevant FAQs (intents) to the nodes will associate that FAQ with the corresponding node name set as a tag.<\/span><\/p>\n

      Here is an example Ontology:\u00a0<\/span><\/p>\n

        \n
      1. Reservation<\/span><\/li>\n
          \n
        1. Make<\/span><\/li>\n
            \n
          1. How do I pay for my booking?<\/span><\/li>\n<\/ol>\n
          2. Update<\/span><\/li>\n
              \n
            1. How do I update my booking?<\/span><\/li>\n<\/ol>\n
            2. Cancel<\/span><\/li>\n
                \n
              1. Can I cancel my reservation?<\/span><\/li>\n<\/ol>\n<\/ol>\n<\/ol>\n

                To create an ontology, follow the below steps:<\/span><\/p>\n

                  \n
                1. Hover over the Knowledge Graph name on the left pane and click <\/span>+<\/strong> icon to <\/span>Add Child Node<\/strong>. For example, we can add a node for questions related to <\/span>international <\/strong>travel, and one for <\/span>reservation <\/strong>related questions.<\/span><\/li>\n
                2. Add.
                  \"\"<\/a><\/span>
                  <\/span><\/li>\n
                3. To add subgroups, hover over the node and use the corresponding <\/span>+<\/strong> icon to add child nodes.
                  \"\"<\/a><\/span>
                  <\/span><\/li>\n
                4. Repeat the process for questions related to\u00a0<\/span><\/li>\n
                5. To add questions to your newly created ontology, select them from your list of FAQs, then drag and drop them onto the relevant node. You will notice the path gets updated accordingly.
                  \"\"<\/a><\/span>
                  <\/span><\/li>\n
                6. Edit a question and see that the node name is added as a tag. This ensures that this question is qualified for a user utterance containing the term\/node name. <\/span>
                  <\/span>
                  \n\n\n\n
                  \n

                  Note <\/em><\/strong>that the tags you added earlier are retained and you can delete them, whereas the node names cannot be deleted from the tags list.<\/span><\/em><\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

                  \"\"<\/a><\/span><\/li>\n

                7. You can further add nodes to your Ontology. The illustration below shows you the Ontology that we will be working with moving forward.
                  \"\"<\/a><\/span>
                  <\/span><\/li>\n
                8. Train the Knowledge Graph and use <\/span>Talk to Bot<\/strong>.<\/span><\/li>\n
                9. There should not be any changes in the responses.<\/span><\/li>\n<\/ol>\n
                  <\/div><\/div>\n

                  Term Usage<\/span><\/span><\/h1>\n

                  By default, the node names or terms are used in evaluating the path and coming up with a set of qualifying questions. You can tighten or relax this rule by either making the term mandatory in the utterance or marking the term\/node as organizer respectively.<\/span><\/p>\n

                  \u00a0<\/p>\n

                  Use Case 1<\/strong>: You want to ensure that the intentions regarding <\/span>Cancel Reservation<\/span><\/em> must be considered only when the user asked a direct question regarding the topic.<\/span><\/p>\n

                  To address this case, we mark the term <\/span>Cancel <\/span><\/em>\u00a0as <\/span>Mandatory<\/strong>. This helps improve the assistant\u2019s performance.<\/span><\/p>\n

                  Steps:<\/span><\/p>\n

                    \n
                  1. Hover over the node, in this case, <\/span>cancel<\/strong>, and click on the <\/span>settings icon<\/strong>.<\/span><\/li>\n
                  2. From the Settings section, select the <\/span>Mandatory<\/strong> option from the dropdown next to the term name.<\/span><\/li>\n
                  3. Save<\/strong> and <\/span>Train<\/strong> the Knowledge Collection.<\/span><\/li>\n
                  4. Now the questions under this node are considered only if the user utterance has the node name, <\/span>cancel <\/span><\/em>in this case.
                    \"\"<\/a><\/span>
                    <\/span><\/li>\n<\/ol>\n

                    \u00a0<\/p>\n

                    Use Case 2<\/strong>: Consider the following FAQs that need to be added to the Knowledge Graph.<\/span><\/p>\n