Amazon Shopping Queries and Products for Ranking Model Training

Amazon shopping queries paired with product information for training product ranking models. The dataset was created by khanmu2003 and was last updated on Hugging Face on December 12, 2025. It contains separate files for query-product pairs with ESCI labels, a product catalog, and source metadata.

Use Cases

• Training product ranking models based on query-product pairs with relevance labels.
• Analyzing query intent and product relevance based on the ESCI label classification system.
• Building or augmenting a product catalog for search applications using the provided product titles and descriptions.

Strengths

• Includes ESCI labels (Exact, Substitute, Complement, Irrelevant) for query-product pairs, which is a structured relevance metric.
• Data is logically separated into distinct files for examples, products, and sources, suggesting organized content.
• Last update timestamp (2025-12-12) is provided, indicating recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, total size, and license information are unknown, which may limit suitability assessment.
• Data may reflect temporal or source bias inherent to the specific Amazon data collection method.

Provenance

Source

Amazon, via Hugging Face user khanmu2003.

Collection Method

Likely gathered from Amazon's shopping platform, but the specific collection method is not detailed.

Time Range

null

Freshness

Last updated 2025-12-12 11:46:49.

Geography

null