Andmekokku

Andmekokku is a term used in Estonian data discourse to denote the process of gathering, integrating, and organizing data from multiple sources into a usable dataset. The goal is to produce a coherent data construct suitable for analysis, reporting, or decision-making. Etymology stems from andmed (data) and kokku (together/assemble). In practice, andmekokku often overlaps with data aggregation and data compilation, though contexts may emphasize different steps.

The process typically includes data discovery, selection, extraction, cleaning, deduplication, normalization, and transformation, followed by enrichment and metadata capture. Data sources may include internal systems, external providers, public datasets, sensor networks, logs, and social feeds. Modern implementations rely on ETL or ELT pipelines, API integrations, database joins, and sometimes web scraping, with governance and quality controls.

Governance aspects cover data quality, lineage, access control, privacy, and compliance with laws such as the GDPR. Privacy-preserving techniques may be used when handling personal data. Data quality dimensions include accuracy, completeness, timeliness, consistency, and validity.

Applications span business analytics, research, regulatory reporting, and machine learning model development. A well-executed andmekokku supports reproducibility, traceability, and scalable data access. It is often distinguished from data storage and from data repositories (andmekogu) by its focus on the preparation and integration of data rather than long-term storage.

See also: andmekogu, ETL, data aggregation.