OTTO Recsys: Real-World E-Commerce Session Data for Multi-Objective Optimization

E-commerce session-based interaction data from OTTO (otto-de) serves as a benchmark for multi-objective optimization in recommender systems. Updated in May 2025, the collection provides real-world retail logs for sequence modeling research.

Use Cases

• Training session-based models to predict user interactions within e-commerce sessions
• Benchmarking multi-objective optimization algorithms for retail recommendation tasks
• Evaluating sequence-based recommendation architectures on real-world OTTO platform logs

Strengths

• Real-world retail data from OTTO
• MIT licensed for open research
• Supports multi-objective optimization benchmarks

Limitations

• Specific column names and schema are not documented in the metadata
• Total record count and file size are unspecified

Provenance

Source

OTTO (otto-de)

Freshness

Updated May 2025