Adaptive Online Learning for Real-Time Spam Filtering in the Music Industry
Bibliographic record
Abstract
Spam detection has evolved significantly with deep learning models such as LSTM networks, resulting in considerable advances in recognizing spam in online environments. However, present solutions are limited to offline training on static datasets and have not been adapted for continuous use in live systems. This study presents an adaptive online learning framework for real-time spam filtering that continuously updates its model with streaming data, addressing the challenge of evolving spam patterns. The proposed framework uses the Keras framework and is deployed on the MusicBrainz platform for the dynamic detection of existing and newly added spam types. Experimental assessments on realworld and synthetic datasets show a significant increase in detection accuracy and adaptability compared to typical batchtrained models. To classify editor accounts in real-time, we created and implemented a REST API that utilizes an LSTM neural network and is continuously enhanced by incorporating fresh feedback. The algorithm continuously improves its accuracy and resilience as it gets knowledge from real-time data and adjusts to new spam trends. This method's practical impact and scalability are demonstrated by its ability to remove 50,000100,000 spam accounts from the MusicBrainz platform and proactively handle future spam threats. In order to increase anti-spam resistance and broaden adaptability, future research will extend this online learning framework to other real-time applications within MetaBrainz, investigating strategies such as online transfer learning.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 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".