bank-transaction

Bank Transaction

Articles:

• Analyze Bank Transaction Data using Graph (Part 1/2)
• Analyze Bank Transaction Data using Graph (Part 2/2)

How to create dataset

Bank transaction simulation dataset for graph analytics.

Pre-created dataset

Sample dataset is under /data/scale-100/ directory.

$ ls /data/scale-100/*.csv
account.csv customer.csv transaction.csv

Copy the 3 CSV files under /data/ for loading.

$ cp ./data/scale-100/*.csv ./data/

Larger dataset (optional)

For creating a graph with larger number of accounts (e.g. 10000), run this script.

$ pip3 install networkx
$ cd script/
$ python3 create-graph.py 10000

This script creates 3 CSV files.

$ ls *.csv
account.csv customer.csv transaction.csv

Locate the CSV files under /data/ directory.

$ mv *.csv ../data/

How to load dataset to tables

Check the service name of the PDB. <connect-string> below is <ip-address>:1521/<service-name>.

$ lsnrctl status

Move to script/ directory.

$ cd script/

Create tables, account, customer, and transaction.

$ sqlplus graphuser/<password>@<connect-string> @create-table.sql

Load the data from the CSV file.

$ sqlldr graphuser/<password>@<connect-string> sqlldr_acc.ctl direct=true
$ sqlldr graphuser/<password>@<connect-string> sqlldr_cst.ctl direct=true
$ sqlldr graphuser/<password>@<connect-string> sqlldr_txn.ctl direct=true