Data Modelling and Import

RDF Insert Methods

Loading RDF Datasets:

QLever supports the ingestion of RDF datasets in various formats, such as Turtle (TTL), RDF/XML, and N-Triples. The process involves:

Specifying the Input Format:

Indicate the RDF format in the configuration or command-line argument.

For example:

qlever index --input-format ttl dataset.ttl

Validation and Parsing:

QLever ensures the integrity of RDF data by validating syntax during ingestion. Adjust parsing parameters, such as locale settings or prefix configurations, to match the dataset specifics​ (https://github.com/ad-freiburg/QLever/blob/master/docs/quickstart.md, https://github.com/ad-freiburg/qlever/discussions/575 ).

Bulk Loading of RDF Datasets

For large RDF datasets, QLever employs optimized bulk-loading mechanisms:

Chunk-based Processing:

Splits datasets into manageable chunks to process them efficiently without exhausting memory.

Partial Vocabulary Construction:

Creates intermediate vocabularies for segments of data, merging them later to complete the indexing process.

Command Example:

qlever index --config qlever_config.json --dataset large_dataset.nt

Here, the configuration file specifies memory and CPU usage limits for the bulk-loading process​.

CSV File Bulk Loader

In addition to RDF, QLever supports importing data from structured CSV files:

Mapping CSV to RDF:

Define how CSV fields map to RDF predicates and subjects using a configuration file or script.

Conversion Tools:

Use external tools (e.g., Python scripts or custom QLever modules) to preprocess the CSV into a compatible RDF format.

Command Example:

qlever csv-import --config csv_config.json data.csv

This method allows users to quickly integrate tabular data into their RDF ecosystem​