Augmenting Document Classification Accuracy Through the Integration of Deep Contextual Embeddings
Bibliographic record
Abstract
Document classification, a fundamental process within the field of natural language processing, has benefitted from the recent advancements in deep learning, particularly in enhancing accuracy.Traditional text clustering methods, such as bag-of-words models, exhibit domain specificity and struggle to handle vast data volumes.They also face limitations in elucidating sophisticated patterns and intricate word and phrase relationships within textual data.These constraints may adversely affect the accuracy of text clustering, subsequently impacting downstream applications like information retrieval, document classification, and natural language processing.This paper proposes a novel text classification model, termed Deep Contextual Embeddings Model (DCEM), designed to improve document classification accuracy.The DCEM integrates pre-trained deep contextual embedding architectures (e.g., GPT-2) with text clustering algorithms (e.g., K-Means).It employs contextual embedding models to enhance document clustering accuracy by capturing context and semantic depth, improving data structure comprehension, and eliminating noise for more precise results.Experimental results, derived from the application of DCEM on AG News, Reuters-21578, and IMDB reviews datasets, indicate a significant improvement in document classification accuracy (81.09%), compared to traditional text clustering and document classification 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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.002 |
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".