{"id":52343,"date":"2023-01-23T07:44:16","date_gmt":"2023-01-23T07:44:16","guid":{"rendered":"https:\/\/developer.kore.ai\/v10-1\/?p=52343"},"modified":"2023-08-21T07:44:01","modified_gmt":"2023-08-21T06:44:01","slug":"answer-from-documents","status":"publish","type":"post","link":"https:\/\/developer.kore.ai\/v10-1\/docs\/bots\/bot-builder-tool\/knowledge-task\/answer-from-documents\/","title":{"rendered":"Answer from Documents (Beta)"},"content":{"rendered":"
\r\n\t
\r\n\t\t
\r\n\t\t\t

The Answer from Documents feature helps answer end-user queries from unstructured PDF documents without the need to extract individual FAQs and train them. This feature leverages a Large Language Model (LLM) and Generative AI models from OpenAI or Azure OpenAI integration to generate answers by processing uploaded documents and user queries. This is a beta feature and is available in English and Non-English languages supported on the XO Platform.<\/span><\/p>\n\n\n\n
Important Note:<\/strong> After redacting personally identifiable information, the documents and end-user queries are shared with OpenAI or Azure OpenAI integration to curate the answers. You must agree to enable the feature.<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
<\/div><\/div>\n

Benefits<\/span><\/span><\/h1>\n
    \n
  1. Automatically identify and answer FAQs using PDF documents as a source of knowledge;<\/span><\/li>\n
  2. Knowledge extraction and VA training are not required.<\/span><\/li>\n<\/ol>\n
    <\/div><\/div>\n

    Important Considerations<\/span><\/span><\/h1>\n
      \n
    1. Answer generation works based on NLU 3.0 and an active integration with OpenAI or Azure OpenAI integration;<\/span><\/li>\n
    2. Answer from Documents acts as a fallback when none of the three NLP engines identifies any intent (using Dialog Tasks or Knowledge AI FAQs);<\/span><\/li>\n
    3. Answers are delivered using a Standard Response with the answer as a dynamic variable.<\/span><\/li>\n<\/ol>\n
      <\/div><\/div>\n

      Limitations<\/span><\/span><\/h1>\n
        \n
      1. You can upload 10 PDF documents, each with a maximum size of 5 MB;<\/span><\/li>\n
      2. Some requests may get rejected, or responses may delay based on OpenAI or Azure OpenAI service availability;<\/span><\/li>\n
      3. This is a beta feature, and requests may take longer to process. We do not recommend it for production use cases as it may directly impact your customers\u2019 experience. Our current recommendation is for VA designers to review generated answers and add them to the Knowledge Graph.<\/span><\/li>\n<\/ol>\n
        <\/div><\/div>\n

        Prerequisites<\/span><\/span><\/h1>\n

        Before using the Answer from Documents feature, you must first perform the following:<\/span><\/p>\n

          \n
        1. Upgrade to NLU 3.0<\/strong> if you have not already done so. You can upgrade by going to <\/span>Natural Language > Advanced Settings<\/strong> and clicking <\/span>Upgrade Now <\/strong>under the <\/span>NLP Version <\/strong>section. \"\"<\/a><\/span><\/li>\n
        2. Configure the OpenAI or Azure OpenAI integration Action<\/strong>. You can find it under <\/span>Integrations > Actions > OpenAI<\/strong>. <\/span>Configuring OpenAI Action<\/span><\/a> or Configuring Azure OpenAI Action<\/a>.<\/span><\/li>\n<\/ol>\n
          <\/div><\/div>\n

          Enable Answer from Documents<\/span><\/span><\/h1>\n

          To enable the Answer from Documents feature, follow these steps:<\/span><\/p>\n

            \n
          1. Go to <\/span>Build > Conversation Skills > Knowledge AI<\/strong>.<\/span><\/li>\n
          2. Under <\/span>Answer from Documents<\/strong>, click <\/span>Get Started<\/strong>. \"\"<\/a><\/span><\/li>\n
          3. To enable the Answer from Documents feature, you must agree to share uploaded documents and user utterances with OpenAI. Please read the contents of the notification window, then check the checkbox next to I agree to share the uploaded documents and user utterances with OpenAI or Azure OpenAI. <\/strong>Click Enable Now<\/strong> to proceed. \"\"<\/a><\/li>\n
          4. Once the feature has been enabled, you can begin uploading PDF documents.<\/span><\/li>\n<\/ol>\n
            <\/div><\/div>\n

            Upload PDF Documents<\/span><\/span><\/h1>\n

            To upload PDF documents, follow these steps:<\/span><\/p>\n

              \n
            1. \n
                \n
              1. Click the <\/span>Upload <\/strong>button that displays after enabling the Answer from Documents feature. \"\"<\/a><\/span><\/li>\n
              2. Drag and drop<\/strong> your document or click <\/span>Browse <\/strong>to find it on your computer.<\/span>
                \n\n\n\n
                Note: Filenames must not be longer than 10 characters.<\/span><\/strong><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

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

              3. Once selecting the document, you can update its name if required. Click <\/span>Proceed <\/strong>when ready.<\/span><\/li>\n<\/ol>\n<\/li>\n<\/ol>\n

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

                  \n
                1. \n
                    \n
                  1. Please wait a few moments until your document uploads. This action may take longer than expected, depending on the file size, your connection speed, and other factors.<\/span><\/li>\n<\/ol>\n<\/li>\n<\/ol>\n

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

                      \n
                    1. \n
                        \n
                      1. When the upload completes successfully, <\/span>close <\/strong>the upload window.<\/span><\/li>\n<\/ol>\n<\/li>\n<\/ol>\n

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

                        Uploaded documents are listed under the <\/span>Answer from Documents<\/strong> section.<\/span> \"\"<\/a><\/p>\n\n\n\n
                        Note<\/strong>: After uploading your first document, the Upload Document option moves to the right side of the section, as shown below. You can use it to add up to 10 PDF documents.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

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

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

                        View Uploaded PDF Documents<\/span><\/span><\/h1>\n

                        To view the contents of an uploaded document, hover over it and click the corresponding <\/span>View <\/strong>icon under <\/span>Actions<\/strong>.\u00a0<\/span> \"\"<\/a>The document will open within a separate window, where you can browse each page, download, print, etc. The viewer is browser-specific. To exit the document view and return to the Knowledge AI section, click the X on the right side corner.<\/span> \"\"<\/a>\u00a0

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

                        Test Answer Generation<\/span><\/span><\/h1>\n

                        A good testing practice for this feature is to ask the VA a question directly related to the contents of your uploaded documents. You can use the <\/span>Talk to Bot <\/span><\/a>feature to test the answers generated from your uploaded documents. The Debug Log within the <\/span>Talk to Bot feature<\/span><\/a> and <\/span>NLP Insights<\/span><\/a> lets you see if a response failure is due to OpenAI or Azure OpenAI timing out.<\/span><\/p>\n

                        You can also run <\/span>Utterance Testing<\/span><\/a> to see results related to use cases where no intent is identified and the query is answered using the <\/span>Answer from Documents<\/span><\/em> feature.<\/span><\/p>\n

                        Batch Testing<\/span><\/em><\/a> and <\/span>Conversation Testing<\/span><\/em><\/a> do not use the <\/span>Answer from Documents<\/span><\/em> feature as a fallback; therefore, no related test results are available.\u00a0<\/span><\/p>\n

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

                        Disable Uploaded PDF Documents<\/span><\/span><\/h1>\n

                        Disabled documents are not considered by the VA when providing answers, even if the Answer From Documents feature is active overall. As such, answers are only provided from uploaded documents that are active.\u00a0<\/span> To disable a document, toggle the <\/span>Status <\/strong>to <\/span>Inactive<\/span><\/em>.\u00a0<\/span> \"\"<\/a>\u00a0

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

                        Delete Uploaded PDF Documents<\/span><\/span><\/h1>\n

                        To delete an uploaded document, hover over it and click the corresponding <\/span>Delete (bin) <\/strong>icon under <\/span>Actions<\/strong>.\u00a0<\/span> \"\"<\/a>You must confirm your choice. Click <\/span>OK <\/strong>to do so, or <\/span>Cancel <\/strong>otherwise.<\/span> \"\"<\/a><\/p>\n\n\n\n
                        Note<\/strong> that you cannot restore deleted documents. You must upload them again if you change your mind later.<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
                        <\/div><\/div>\n

                        Disable Answer from Documents<\/span><\/span><\/h1>\n

                        If you disable Answer from Documents, queries will no longer be sent to LLMs as a fallback, but documents are stored in the database, and you can enable the feature again later.<\/span> To disable the feature<\/span>, turn off the toggle <\/strong>at the far right of the section.<\/span> \"\"<\/a>You must confirm your choice. Click <\/span>Confirm <\/strong>to do so or <\/span>Cancel <\/strong>otherwise.<\/span> \"\"<\/a><\/p>\n

                        Your documents stay stored while the feature is disabled. You can re-enable it any time later by turning on the same toggle.<\/span> \"\"<\/a><\/p>\n

                        <\/div><\/div>\n\r\n\t\t<\/div>\r\n\t<\/div>\r\n<\/div><\/div><\/div><\/div><\/div><\/section>\n","protected":false},"excerpt":{"rendered":"The Answer from Documents feature helps answer end-user queries from unstructured PDF documents without the need to extract individual FAQs and train them. This feature leverages a Large Language Model (LLM) and Generative AI models from OpenAI or Azure OpenAI integration to generate answers by processing uploaded documents and user queries. This is a beta...","protected":false},"author":26,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[33],"tags":[],"class_list":["post-52343","post","type-post","status-publish","format-standard","hentry","category-knowledge-task"],"_links":{"self":[{"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/posts\/52343","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/users\/26"}],"replies":[{"embeddable":true,"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/comments?post=52343"}],"version-history":[{"count":19,"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/posts\/52343\/revisions"}],"predecessor-version":[{"id":58870,"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/posts\/52343\/revisions\/58870"}],"wp:attachment":[{"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/media?parent=52343"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/categories?post=52343"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/developer.kore.ai\/v10-1\/wp-json\/wp\/v2\/tags?post=52343"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}