Using Machine Learning Techniques to Improve the Accuracy of Computer Translation in the English Translation of Specialized Terms
Bibliographic record
Abstract
At present, machine translation performs better in the general domain translation effect of large-scale bilingual corpus, but the translation effect in specific domains still needs to be improved.In order to optimize the accuracy of machine translation in the domain of English translation of professional terms, this paper proposes a translation model that incorporates syntactic knowledge and terminology.Aiming at the problem of more limited translation domain knowledge in the RNMT and Transformer models based on the self-attention mechanism, an optimization method is proposed.According to the domain characteristics of English translation of professional terms, English syntactic keywords are incorporated into the model training process, the special information contained inside the text of professional terms is learned, and the lexical properties of each word in the dataset are recognized before they are input into the model.Then attempts are made to incorporate the specialized terminology into the model to enrich the parallel corpus required by the model.The experiments confirm the excellent performance of the optimized translation model in this paper on the De→En terminology translation task, which improves 22.67 BLEU values compared to the base model.And the fluctuation of its BLEU value with the change of sentence length is small, which further indicates that the method optimizes the accuracy of the machine translation model in the English translation of professional terms.
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 | 0.003 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| 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 | 0.002 |
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".