mapper-ofac

Overview

The ofac_mapper.py python script converts the US Treasury's OFAC SDN list into a json file ready to load into senzing. You will first need to download the sdn.xml file from here .

Loading watch lists requires some special features and configurations of Senzing. These are contained in the ofac_config_updates.g2c file.

Usage:

usage: ofac_mapper.py [-h] [-i INPUTFILE] [-o OUTPUTFILE] [-l LOGFILE]

optional arguments:
  -h, --help            show this help message and exit
  -i INPUTFILE, --inputFile INPUTFILE
                        an sdn.xml file downloaded from https://www.treasury.gov/ofac/downloads.
  -o OUTPUTFILE, --outputFile OUTPUTFILE
                        output filename, defaults to input file name with a .json extension.
  -l LOGFILE, --logFile LOGFILE
                        optional statistics filename in json format.

Contents

1. Prerequisites
2. Installation
3. Running the ofac_mapper mapper
4. Configuring Senzing
5. Loading into Senzing
6. Optional ini file parameter

Prerequisites

• python 3.6 or higher
• Senzing version 2.0 or higher

Installation

Place the the following files on a directory of your choice ...

• ofac_mapper.py
• ofac_config_updates.g2c
• ofac_codes.csv

Note: The ofac_codes.csv will grow through time as new codes are added by them. These new codes will need to be mapped to their Senzing equivalents. New idTypes will need to be mapped to national_id, other_id, etc. New idCountries will need to be mapped to their 3 character iso equivalent. A good way to detect new codes is to run the ofac_mapper against the latest file and then check the ofac_codes.csv for any codes that have not been reviewed. Once you review and update them, run the mapper a second time to pick up your updates.

Running the ofac_mapper mapper

First, download the latest sdn.xml file from https://www.treasury.gov/ofac/downloads. This is the only file needed. It is a good practice to rename it based on the publish date such as sdn-yyyy-mm-dd.xml and place it on a directory where you will store other source data files loaded into Senzing.

Second, run the mapper. Typical usage:

python ofac_mapper.py -i /<path-to-file>/sdn-yyyy-mm-dd.xml -o /<path-to-file>/sdn-yyyy-mm-dd.json -l mapping_stats.json

Note The mapping satistics should be reviewed occasionally to determine if there are other values that can be mapped to new features. Check the UNKNOWN_ID section for values that you may get from other data sources that you would like to make into their own features. Most of these values were not mapped because there just aren't enough of them to matter and/or you are not likely to get them from any other data sources. However, DUNS_NUMBER, GENDER, and WEBSITE_ADDRESS were found by reviewing these statistics!

Configuring Senzing

Note: This only needs to be performed one time! In fact you may want to add these configuration updates to a master configuration file for all your data sources.

From your Senzing project's python directory, run ...

python3 G2ConfigTool.py <path-to-file>/ofac_config_updates.g2c

This will step you through the process of adding the data sources, features, attributes and other settings needed to load this watch list data into Senzing. After each command you will see a status message saying "success" or "already exists". For instance, if you run the script twice, the second time through they will all say "already exists" which is OK.

WARNING: The are a few commented out optional settings described in the configuration file as they affect performance and quality. Only use them if you understand and are OK with the effects.

Loading into Senzing

If you use the G2Loader program to load your data, from your Senzing project's python directory, run ...

python G2Loader.py -f /<path-to-file>/ofac-yyyy-mm-dd.json

Optional ini file parameter

There is also an ini file change that can benefit watch list matching. In the pipeline section of the main g2 ini file you use, such as the /opt/senzing/g2/python/G2Module.ini, place the following entry in the pipeline section as show below.

[pipeline]
 NAME_EFEAT_WATCHLIST_MODE=Y

This effectively doubles the number of name hashes created which improves the chances of finding a match at the cost of performance. Consider creating a separate g2 ini file used just for searching and include this parameter. If you include it during the loading of data, only have it on while loading the watch list as the load time will actually more than double!