Detect Fraud in Payment Streams | Dataford Interview Questions

Write a function to detect fraudulent transaction patterns in a stream of payment data.

Implement detect_fraudulent_patterns(transactions, window_seconds, max_transactions). Each transaction is a dictionary with account_id and integer timestamp, and timestamps are in nondecreasing order. Return a sorted list of account IDs that have more than max_transactions transactions within any trailing window_seconds interval. An account should appear only once in the result.

Constraints

0 <= transactions.length <= 5000
Each transaction contains an account_id string and a nonnegative integer timestamp
Transactions are ordered by nondecreasing timestamp
1 <= window_seconds
1 <= max_transactions
Return account IDs in lexicographic order

Function Signature

def detect_fraudulent_patterns(transactions, window_seconds, max_transactions):

Your question is Detect Fraud in Payment Streams. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

Write a function to detect fraudulent transaction patterns in a stream of payment data.

Constraints

0 <= transactions.length <= 5000
Each transaction contains an account_id string and a nonnegative integer timestamp
Transactions are ordered by nondecreasing timestamp
1 <= window_seconds
1 <= max_transactions
Return account IDs in lexicographic order

Function Signature

def detect_fraudulent_patterns(transactions, window_seconds, max_transactions):

Your question is Detect Fraud in Payment Streams. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.