As Moveworks builds and integration with LumApps, this guide highlights the capabilities that can be built with this integration and the end user experience while fetching knowledge articles.

Moveworks utilizes LumApps' APIs to discover LumApps’ response structure and how to integrate those responses into the Moveworks knowledge ingestion engine.

Lumapps Knowledge Structure

How do we ingest knowledge articles?

Our knowledge engine makes relevant snippets from the knowledge articles in less than a second, using Moveworks' natural language understanding and similarity scoring.

We store the snippzetized articles in our enterprise cache, which can be served to users based on the inquiry.

We are currently fetching all the pages along with its meta data such as URL, Id, Author details, Status using the below API.

URL

https://go-cell-002.api.lumapps.com/_ah/api/lumsites/v1/content/list?lang=en

Integration Scope

LumApps hosts knowledge under different pages and sub-pages, which acts a container that hold different widgets.

These are some widgets types available on the LumApps platform: HTML, introduction text, image, video, comments, user lists, etc.

As LumApps doesn’t return responses in an HTML format, we create an HTML file and after analyzing the response, we can only ingest the intro and HTML widget content into the created HTML file.

So we ingest only:

1. Intro widget
2. HTML widget

After creating an HTML file we store it in our cache, our knowledge ingestion engine works on identifying HTML tags and creating relevant knowledge base article snippets.

End user experience

Whenever a user makes an inquiry there are two cases which can occur:

Case 1: Bot finds a relevant knowledge article

Case 2: Bot doesn’t find any relevant knowledge article