This is an interesting use case, especially when user behavior is better represented as a sequence rather than a single aggregated vector. Treating events as ordered inputs and encoding them with a transformer or RNN-style approach can produce much richer user embeddings. A somewhat related discussion is here: https://huggingface.co/proxy/discuss.huggingface.co/t/predict-next-embedding-given-sequence-of-embeddings/31845
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Combine multiple embeddings from different authors | 0 | 184 | May 15, 2024 | |
| Sequence analysis | 0 | 239 | October 3, 2020 | |
| Seeking Advice🔥🔥| Strategy for Embedding Multiple Subjective Reviews in One-time Event Domain Recommendations | 1 | 62 | January 2, 2026 | |
| Train the Best Sentence Embedding Model Ever with 1B Training Pairs | 36 | 27853 | July 2, 2023 | |
| Predict next embedding given sequence of embeddings | 0 | 419 | February 16, 2023 |