Pre-processed document upload information

Epiq AI can ingest documents that are previously processed and exported using the Concordance DAT file format.

Folder structure

The preprocessed documents should be arranged in different volume folders. Each volume contains the following:

a DATA folder containing metadata load file(s), in .dat or .csv
a TEXT folder containing text files
Optional: a NATIVES folder containing native files

Note: The Load file(s) can be put outside of the DATA folder but need to be at the same level as the TEXT folder.

Load file format

Epiq AI follows the Load file specifications used by Relativity.

Fields

Epiq AI requires your load file to contain a Document ID field as a unique identifier for a document. Epiq AI automatically maps your input fields to Epiq AI’s internal fields. You can choose to manually map a field or not import a field. You can choose to import a field that does not map to an existing Epiq AI field. Epiq AI will automatically create a custom field using its original naming. Below is a list of Epiq AI fields:

Epiq AI Field	Field Type	Description
DOC_ID	Text	Document ID
PARENT_ID	Text	Parent document ID
CHILDREN_IDS	Text	A list of chilldren document IDs separated by semicolon
FIRST_BATES	Text	First Bates associated with the document
LAST_BATES	Text	Last Bates associated with the document
BEG_ATTACH	Text	Parent document ID or the beginning bates of the family group
END_ATTACH	Text	The last attachment or the end bates of the family group
IS_PARENT	Yes/No	Whether the document is a parent in the family.
FROM	Text	Sender of the email.
TO	Text	Recipient(s) of the email.
CC	Text	Recipient(s) of the email.
BCC	Text	Recipient(s) of the email.
FROM_SMTP	Text	Sender of the email, in the SMTP format.
TO_SMTP	Text	Recipient(s) of the email, in the SMTP format.
CC_SMTP	Text	Recipient(s) of the email, in the SMTP format.
BCC_SMTP	Text	Recipient(s) of the email, in the SMTP format.
FROM_DOMAIN	Text	Sender domain.
TO_DOMAIN	Text	Recipient domain(s).
CC_DOMAIN	Text	Recipient domain(s).
BCC_DOMAIN	Text	Recipient domain(s).
SUBJECT	Text	Subject of the email.
DATE_SENT	DateTime	Date Sent of the email (date and time combined in one field).
DATE_RECEIVED	DateTime	Date Received of the email (date and time combined in one field).
DATE_CREATED	DateTime	File system created date (date and time combined in one field).
DATE_MODIFIED	DateTime	File system last modified date (date and time combined in one field).
FILENAME	Text	Original file name of the file.
FILE_SIZE	Number	Size of the original file in bytes.
FILE_EXTENSION	Text	Extension of the file.
ORIG_FILE_PATH	Text	Original path of the associated file with the file name.
TITLE	Text	Title of the efile extracted from the file’s metadata.
COMPANY	Text	Company of the efile extracted from the file’s metadata.
AUTHORS	Text	Author(s) of the efile extracted from the file’s metadata.
COMMENTS	Text	Comments of the efile extracted from the file’s metadata.
CUSTODIAN	Text	Custodian name.
ALL_CUSTODIANS	Text	Custodian name(s) for the master file and the duplicate copies of the file.
HASH	Text	The Hash of the file.
LANGUAGES	Text	The languages identified within the file
IMPORTANCE	Text	Priority of the email.
SENSITIVITY	Text	The Sensitivity of the email is the level of importance for an email.
IS_ENCRYPTED	Yes/No	Whether the file is encrypted.
CHANNEL_NAME	Text	Channel name/ID for chat.
PARTICIPANTS	Text	Active participants in chat.
CHAT_START_TIME	DateTime	Chat start time.
CHAT_END_TIME	DateTime	Chat end time.
RSMF_DOC_ID	Text	RSMFDocID exported by Relativity.

Overview

Fundamentals

AI Chat

Work with projects

Release Notes

Product Guides

System

Use Ask AI

Change logs

Sandbox

Troubleshooting

Internal

Epiq AI V2

Pre-processed document upload information

Folder structure

Load file format

Fields

​Folder structure

​Load file format

​Fields

Folder structure

Load file format

Fields