AWS · MLA-C02 · BETA
Data quality, labeling, cleaning, and masking
MLA-C02 · Lesson 5 of 23 · 32:11
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 5 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 1, data preparation). In this lesson: data quality rules with AWS Glue Data Quality and DataBrew, labeling and annotation, outliers, missing values, and duplicates, and masking, redaction, and anonymization of sensitive data. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- Each night, a Glue ETL job loads new trips. It must stop when trip IDs repeat or durations are negative, and the team wants a warning when the volume drops oddly, without writing every rule. What should the ML engineer use?
- A few payment amounts are missing, and the distribution has extreme values from legitimate corporate rentals. How should the ML engineer fill the missing amounts?
- Support call transcripts will train a model that classifies complaints, and they contain names, phone numbers, and card numbers in free text. What should the ML engineer do before training?