Scalable Distributed kNN Processing on Clustered Data Streams
Bibliographic record
Abstract
Recommender systems provide an important tool for users to find interested items from the massive amount of user-generated contents. As user interests often change over time and contents become available in a streaming fashion, it is highly desirable to support real-time recommendation that can adapt to changes in user interests and contents. If we represent both user interests and items by high-dimensional points in the same vector space, we can recommend to the user the $k$ items that are the nearest neighbors (kNN) of the user. The problem of real-time recommendation, thus, translates to computing the kNNs based on the most recent items when the user interests change. As such, the main issue we tackle in this paper is to efficiently process high-dimensional kNN queries over a sliding window on data streams. In particular, we are interested in developing a scalable distributed solution to be able to handle the ever-increasing number of users and volume of data. We propose a new index structure called the dynamic bounded rings index (DBRI) to index the data points in data streams. The basic idea is to first find a set of pivots and assign all points to their nearest pivot to form subsets and then partition each subset into finer-grained bounded rings that can be dynamically adjusted as points change. The design of DBRI lends itself to easy adoption in a distributed setting. We further present the distributed high-dimensional kNN query algorithm (DHDKNN) based on DBRI, aiming at reducing both the communication and the computational cost of query processing. The experiments demonstrate that our algorithm scales well and significantly outperforms the existing methods.
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 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".