Ingesting Unstructured Data

Custom Ingestion without creating an AI application. This is for users who want to create their own custom data flows on the extracted PDF or website data.

Access Ingest Settings:

• Click the 'Ingest' link on the AI Applications page.

Create an ingest Job:

• Press the '+' button and name your ingestion job.

Configuring Sources

• PDF Files:

  • For PDF, Select PDF files

  • You can upload PDFS from your own device or an object storage like GCS or S3

  
• Website URL(HTML):

  • For URL links, select the file type - html.

  • Add URLs of interest, check the sub-crawling/javascript enabled boxes, and configure the custom crawler settings if interested.

  

Submit Ingestion:

• Once finished, choose the ingest target location and hit submit to perform the ingestion.