If you try to load data from S3 to NEO4J you are going to need to presing the file. So you need to expose the data to somebody that have the file.

So, first you need to presing the file:

aws s3 presign s3://rohank/actors.csv

And then you can load the data on your NEO4J Instance:

LOAD CSV WITH HEADERS FROM "https://rohank.s3.amazonaws.com/actors.csv?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAICM6A3RO53KOKSRQ%2F20190404%2Fus-east-2%2Fs3%2Faws4_request&X-Amz-Date=20190404T215301Z&X-Amz-Expires=3600&X-Amz-SignedHeaders=host&X-Amz-Signature=61cb485af12daa60bb8cb7a91fb503797311c8e178d9bfa3c7ff49770e4535b5" as row return count(row)

If you need to load big amount of data, and you don’t want to expose your information you need to think to use another kind of way to import the information. For example using data-import or adding the script that expose the information to a VPC