Long- and Short- Term Sequential Recommendation with Temporal Interval
Bibliographic record
Abstract
Sequential recommendation aims to learn the changes of users’ interests according to their historical behaviors and predict the most likely next item. Since user’s historical behavior are sequential actions, user’s interest in different time periods has different emphases. For predicting a user’s next item, not only the recent behavior is important, but also long-term preference with all historical behaviors could not be ignored. Recently, self-attention based user preference modeling has drawn much attention for its advantages of fewer parameters and parallelism. However, most of the existing self-attention based model do not make a good distinction between the long-term and short-term preferences of users. Based on the observations, we design a sequential recommendation model based on a combination of long-term and short-term preference. On the one hand, considering the changes of users’ interests in different periods, we divide the sequence of users’ behaviors into different temporal windows. Then, we use GRU to capture users’ interests in different temporal window. On the other hand, to better combine the global and local information of user’s, we adopt a locally constrained multi-head attention mechanism based on Transformer encoder. On three real-world public datasets, we finally validate the efficacy of our method.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".