Robility Knowledge Ingestion

Estimated reading: 2 minutes

The Robility Knowledge Ingestion component is designed to create and manage a knowledge base that serves as a central repository of factual and contextual information. This knowledge base can include structured data, documents, FAQs, or other reference materials that enhance automation processes by enabling intelligent decision-making and context-aware interactions.

This component allows users to define, embed, and store knowledge for retrieval and use in AI-powered automation, analytics, and conversational workflows within Robility Flow.

How It Works

1. Create a Knowledge Base

a. Select the Knowledge parameter and choose “Create new knowledge list” from the dropdown.
b. Provide a unique name for the knowledge base.
c. Choose an embedding provider and configure the corresponding embedding API key.

2. Add Input Data

a. Add the input file containing your knowledge data using the Search Files from the file system component.
b. The input must be of type data or dataframe, which should be pre-processed or chunked before ingestion.
c. The data will be processed and embedded into the knowledge base for retrieval during automation.

Parameters

Parameter	Description
Knowledge	Name of the knowledge base to be created or updated.
Input	Data or DataFrame input containing the knowledge content to be ingested. The data must already be chunked or processed.
Column Configuration	Defines which columns from the input data should be used for processing and embedding.
Chunk Size	Specifies the data chunk size used during ingestion for optimized embedding and retrieval. By default, the value is set to 1000.
Embedding Provider API Key	API key for the selected embedding provider used to generate vector representations of the knowledge data.
Allow Duplicates	Indicating whether duplicate data entries should be allowed during ingestion.

Robility Knowledge Ingestion

How It Works

Parameters

Robility Knowledge Ingestion

CONTENTS