Mobile Money#
Example file: synthetic_data/mobilemoney.csv
txn_type |
caller_id |
recipient_id |
timestamp |
amount |
sender_balance_before |
sender_balance_after |
recipient_balance_before |
recipient_balance_after |
|---|---|---|---|---|---|---|---|---|
cashin |
jb9f3j |
2020-01-01 00:00:42 |
234.0 |
500.0 |
734.0 |
|||
cashout |
ad8b3b |
2020-02-12 10:12:08 |
195.0 |
200.5 |
5.5 |
|||
p2p |
oewf9ras |
bncz83 |
2020-03-30 14:50:01 |
2.4 |
100.0 |
97.6 |
78.1 |
80.5 |
billpay |
bncz83 |
jb9f3j |
2020-06-15 22:38:31 |
76.1 |
150.0 |
73.9 |
12.9 |
89.0 |
other |
bncz83 |
ad8b3b |
2020-08-16 23:10:10 |
502.0 |
872.0 |
370.0 |
0.00 |
502.0 |
Fields
#
*txn_type: string Any string describing the type of the transaction; the dataset can define any number of categories for transactions, but keep in mind that features will be calculated separately for each transaction type so compute time scales with the number of categories. Common categories include {cashin, cashout, billpay, p2p, other}.
*caller_id: string
Unique ID for the caller or sender
*recipient_id: string
Unique ID for the recipient
*timestamp: string
String in format ‘YYYY-MM-DD hh:mm:ss’ representing the start time of a call or the time an text is sent
*amount: float
Amount of money transferred (in any currency)
sender_balance_before: float
Balance of sender before transaction (in any currency). Only provided by some mobile network operators.
sender_balance_after: float
Balance of sender after transaction (in any currency). Only provided by some mobile network operators.
recipient_balance_after: float
Balance of the recipient after transaction, where applicable (in any currency). Only provided by some mobile network
operators.
recipient_balance_before: float
Balance of the recipient after transaction, where applicable (in any currency). Only provided by some mobile network
operators.
Note
Columns without a preceding asterisk ‘*’ are optional