Researchers release SimTrace for generating synthetic user behavior data
Researchers introduced SimTrace, an open-source framework that creates synthetic, multimodal user interaction data for online modeling. The tool simulates fine-grained clickstreams by anonymizing real user behavior and replicating web environments, addressing gaps in existing datasets that either lack detail or are platform-specific.
Key points
- SimTrace generates synthetic user trajectories from anonymized real interactions and web environments
- Outperforms baselines on 7 of 8 fidelity metrics in e-commerce testing
- Augmenting real data with synthetic improves next-action prediction by 11.0%
SimTrace outperformed competitors on 7 of 8 fidelity metrics and improved next-action prediction accuracy by 11.0% when combined with real data. The authors tested it in an e-commerce context, showing synthetic data can train models as effectively as real data for tasks like purchase prediction and recommendations. The project is available open-source to support research in user behavior modeling.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- arXiv study tests AI agent on rediscovering Blaschke-curve invariant · 1 src
- Researchers introduce GFlowNets for diverse synthetic expert conversations · 1 src
- arXiv study finds AI tutors’ guidance varies by impasse type · 1 src
- CARAT study finds materials LLMs often recite rather than reason about crystal structures · 1 src
- Researchers find AI agents can radicalize each other in simulated conversations · 1 src
Comments
via GitHub Discussions