The Article summarizer is one of the many Artificial Intelligence (AI) powered features in Document360. This feature comes in handy when readers want to save time by not reading lengthy articles. Instead, they can read a piece of condensed article information as a summary.

Document360 uses natural language processing (NLP) to locate the vital information in the article content while maintaining the original context to generate an optimal summary.


Prerequisites

  • Preprocessed article content should be more than 250 words
  • The AI bundle addon is already purchased

The article summary is available (depending on your subscription plan) in new/existing articles published on or after May 27, 2023.


Enabling the 'Summary' section in articles

To enable the summary section, follow the below steps

  1. From the Knowledge base portal, click on

Settings > Knowledge base site > Article settings & SEO > Article settings(tab)

  1. Expand the Article header section
  2. Turn on the Show AI article summarizer toggle 'On'

By default, the summary option is disabled (toggle 'Off') for all new and existing projects

When enabled, the Summary is available at the article top in the Knowledge base site. Click on 'Summary' to expand and read it.


Credits usage

  • When a summary is generated for an article, one credit is consumed
  • The credit limit is assigned is standard in the addon bundle
  • The credit provided does not renew as a monthly limit
  • The credits can be increased on-demand by contacting customer support

What is 'Preprocess' or 'Preprocessed words' in Document360 articles?

Preprocess refers to filtering specific article/category page elements. The below elements are not considered for meta description generation's word count:

  • HTML tags
  • Images
  • URL/links
  • Code blocks

    If you have an article word count shown as 220 words at the bottom of the Document360 editor, this includes the HTML tags, links, code blocks, and more. When preprocessing content is considered, the word count would be less. Ensure the preprocessed article content is more than 50 words for the title recommender to work.


Data privacy

We understand that our customer's Knowledge base data is sensitive. As this feature uses a form of OpenAI integration, we adhere to the privacy policies on OpenAI. An extract from the policy states, "OpenAI will not use data submitted by customers via our API to train OpenAI models or improve OpenAI's service offerings."

Read the complete draft of OpenAI API data usage policies

If you any queries regarding Document360's data policy, please read our Privacy policy