TriDeepRec: A Hybrid Deep Learning Approach to Content and Behaviour-based Recommendation Systems
Bibliographic record
Abstract
Abstract Hybrid recommendation systems are increasingly crucial for businesses aiming to boost revenue and customer engagement. These systems integrate various algorithms, each with unique strengths, to outperform traditional recommendation methods. Our study introduces a novel hybrid recommendation system, TriDeepRec, which effectively combines content-based and behaviour-based data to enhance recommendation accuracy. We first introduce a Convolutional Autoencoder-based Recommendation System (CAERS), designed to process content data and extract complex, meaningful patterns, translating these into predictive ratings. Notably, CAERS tackles the cold-start problem by leveraging content information alone, making it robust in scenarios where historical user interaction data is sparse or unavailable. Next, we incorporate Neural Collaborative Filtering (NCF), a deep learning approach, to analyze past user behaviour and predict ratings. The outputs from CAERS and NCF are then integrated using a Multilayer Perceptron (MLP), a type of neural network, to generate the final recommendations. Our methodology employs three deep learning techniques to create TriDeepRec, a system capable of utilizing both past interactions and content attributes. We evaluate our system using two datasets, initially focusing on CAERS to demonstrate its effectiveness in addressing the cold-start problem. Subsequently, we assess the performance of TriDeepRec as a whole. The results, measured in terms of Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), indicate significant improvements over both the individual components and other leading models in the field. This demonstrates that TriDeepRec, by 1 harnessing the strengths of both content and behaviour data, provides a more accurate and reliable recommendation system.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".