Dataford
Interview QuestionsInterview GuidesExperiencesMock InterviewsPricing
Get started
SQL and Excel Anomaly Detection
00:00
5 left

SQL and Excel Anomaly Detection

MediumSQL · PostgreSQL

Problem

How would you use SQL or advanced Excel to identify anomalies in a dataset containing millions of transit records?

Using the provided transit records and route reference data, identify records with invalid values, unusually long or short durations, fares above the route maximum, duplicate trip signatures, missing route references, or timestamps earlier than the preceding vehicle sequence.

Output

  1. One row per anomalous record, with record_id, anomaly_score, anomaly_types, route_id, vehicle_id, and boarded_at.
  2. Include only records with at least one anomaly, ordered by descending score, then boarded_at ascending and record_id ascending.

Schema

transit_records
ColumnTypeDescription
record_idPKINTUnique transit record identifier
vehicle_idVARCHAR(20)Vehicle identifier
route_idVARCHAR(20)Recorded route identifier
trip_sequenceINTSequence number for the vehicle
boarded_atTIMESTAMPBoarding timestamp
card_idVARCHAR(30)Transit card identifier
origin_stationVARCHAR(50)Origin station
destination_stationVARCHAR(50)Destination station
duration_minutesDECIMAL(8,2)Recorded trip duration
fare_amountDECIMAL(8,2)Charged fare amount
route_reference
ColumnTypeDescription
route_idPKVARCHAR(20)Reference route identifier
expected_duration_minutesDECIMAL(8,2)Expected route duration
max_fareDECIMAL(8,2)Maximum valid fare for the route
Tablestransit_recordsroute_reference
Interviewer

Your question is SQL and Excel Anomaly Detection. Start with the requirements and the two tables 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.

You need to log in / sign up to run or submit.
CodePostgreSQL
You need to log in / sign up to run or submit.Ln 1
Run your query to see results here.