From 89cf9c17882717d72f515ecf0c6d9e5a4a450ad9 Mon Sep 17 00:00:00 2001
From: rderbier <rderbier@gmail.com>
Date: Tue, 14 Jan 2025 16:52:42 -0800
Subject: [PATCH] add IBM Anti Money Laundering (aml) dataset

---
 data/aml/AML_Trans.schema        | 37 ++++++++++++++++++++++++++++++++
 data/aml/HI-Small_Trans.rdf.gz   |  3 +++
 data/aml/HI-Small_Trans.template | 23 ++++++++++++++++++++
 data/aml/README.md               | 14 ++++++++++++
 4 files changed, 77 insertions(+)
 create mode 100644 data/aml/AML_Trans.schema
 create mode 100644 data/aml/HI-Small_Trans.rdf.gz
 create mode 100644 data/aml/HI-Small_Trans.template
 create mode 100644 data/aml/README.md

diff --git a/data/aml/AML_Trans.schema b/data/aml/AML_Trans.schema
new file mode 100644
index 0000000..b60382f
--- /dev/null
+++ b/data/aml/AML_Trans.schema
@@ -0,0 +1,37 @@
+<Account.bank>: uid .
+<Account.id>: string @index(hash) .
+<Bank.id>: string @index(hash) .
+<Transaction.amount_paid>: float  @index(float) .
+<Transaction.amount_received>: float   @index(float) .
+<Transaction.from>: uid @reverse .
+<Transaction.payment_currency>: string .
+<Transaction.payment_type>: string  @index(hash) .
+<Transaction.receiving_currency>: string .
+<Transaction.to>: uid @reverse .
+<Transaction.timestamp>: datetime @index(day) .
+<Transaction.laundering>:  bool @index(bool) .
+<Transaction.id>: string @index(hash) .
+
+<xid>: string @index(exact) @upsert .
+
+type <Bank> {
+	<Bank.id>
+}
+
+type <Account> {
+    <Account.id>
+    <Account.bank>
+}
+
+type <Transaction> {
+    <Transaction.id>
+    <Transaction.timestamp>
+    <Transaction.amount_paid>
+    <Transaction.amount_received>
+    <Transaction.from>
+    <Transaction.payment_currency>
+    <Transaction.payment_type>
+    <Transaction.receiving_currency>
+    <Transaction.to>
+    <Transaction.laundering>
+}
diff --git a/data/aml/HI-Small_Trans.rdf.gz b/data/aml/HI-Small_Trans.rdf.gz
new file mode 100644
index 0000000..a7bc10c
--- /dev/null
+++ b/data/aml/HI-Small_Trans.rdf.gz
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66a9f0656a50495bd268d03dccd0f1df4523a30eeca94129db4cb8ce1469c908
+size 164364288
diff --git a/data/aml/HI-Small_Trans.template b/data/aml/HI-Small_Trans.template
new file mode 100644
index 0000000..519d1a1
--- /dev/null
+++ b/data/aml/HI-Small_Trans.template
@@ -0,0 +1,23 @@
+# Timestamp,From Bank,Account,To Bank,Account,Amount Received,Receiving Currency,Amount Paid,Payment Currency,Payment Format,Is Laundering
+
+<_:Bank_[From Bank]> <dgraph.type> "Bank" .
+<_:Bank_[From Bank]> <Bank.id> "[From Bank]" .
+<_:Bank_[To Bank]> <dgraph.type> "Bank" .
+<_:Bank_[To Bank]> <Bank.id> "[To Bank]" .
+<_:A_[From Account]> <dgraph.type> "Account" .
+<_:A_[From Account]> <Account.id> "[From Account]" .
+<_:A_[From Account]> <Account.bank> <_:Bank_[From Bank]> .
+<_:A_[To Account]> <dgraph.type> "Account" .
+<_:A_[To Account]> <Account.id> "[To Account]" .
+<_:A_[To Account]> <Account.bank> <_:Bank_[To Bank]> .
+<_:T_[LINENUMBER]> <Transaction.id> "[Timestamp],[From Bank],[From Account]" .
+<_:T_[LINENUMBER]> <dgraph.type> "Transaction" .
+<_:T_[LINENUMBER]> <Transaction.from> <_:A_[From Account]> .
+<_:T_[LINENUMBER]> <Transaction.to> <_:A_[To Account]> .
+<_:T_[LINENUMBER]> <Transaction.timestamp> "=datetime([Timestamp],%Y/%m/%d %H:%M)" .
+<_:T_[LINENUMBER]> <Transaction.amount_received> "[Amount Received]" .
+<_:T_[LINENUMBER]> <Transaction.receiving_currency> "[Receiving Currency]" .
+<_:T_[LINENUMBER]> <Transaction.amount_paid> "[Amount Paid]" .
+<_:T_[LINENUMBER]> <Transaction.payment_currency> "[Payment Currency]" .
+<_:T_[LINENUMBER]> <Transaction.payment_type> "[Payment Format]" .
+<_:T_[LINENUMBER]> <Transaction.laundering> "[Is Laundering]" .
diff --git a/data/aml/README.md b/data/aml/README.md
new file mode 100644
index 0000000..22663d3
--- /dev/null
+++ b/data/aml/README.md
@@ -0,0 +1,14 @@
+
+IBM Transactions for Anti Money Laundering (AML)
+from 
+https://www.kaggle.com/datasets/ealtman2019/ibm-transactions-for-anti-money-laundering-aml
+
+The folder contains a schema file and RDF files created from 
+HI-Small_Trans.csv
+
+RDF file has been created using [csv_to_rdf](https://github.com/hypermodeinc/dgraph-experimental/blob/main/data-import/csv-to-rdf/csv_to_rdf.py) python script from 
+dgraph experimental repository, using the template file provided in this folder.
+
+See the corresponding HI_Small_Patterns.tx file for generated fraudulent transactions. These transactions are labeled as Transaction.laundering true in the dataset.
+ 
+