A Study of Word Vector Computation and Multilayer Network Representation in English Corpus
Bibliographic record
Abstract
Although China's research on English is not as early as that of the western countries, researchers, combining the basic national conditions of China and the actual situation of the nationals' learning of English, have been making continuous efforts in the research on the construction and application of English corpus, and have already achieved satisfactory results.In this paper, we first analyze the related contents of English corpus, and construct English corpus corpus from phonological and semantic aspects by analyzing the correlation characteristics between English corpus and semantics, according to the basic principles of corpus selection.Combining two word vector similarity measures, Jaccard similarity and edit distance, finally constitutes the final similarity calculation algorithm for English sentences.The MECNC model is constructed by integrating the joint representation and co-representation learning methods, and using edge probability to abstract the connection between two nodes.Experimentally analyze the word vector similarity of English corpus with the results of English corpus recommendation based on multilayer network representation.The correlation scores of Jaccard similarity metric in WS-SIM, WS-REL, MEN, Mtruk-771, and Simverb-3500 are 0.8069, 0.6668, 0.7389, 0.7125, respectively, 0.2769, which achieves the best results, so Jaccard captures more of the correlation between words.Experiments on link prediction task were conducted on five corpora using 3, 5, 8, and 10-fold cross-validation methods, and on the corpus CKM [245,1550], MECNC model OM3 has a maximum AUC value close to 0.94 at a cross-validation number of 8, which shows that MECNC, which is used as a guiding information for intra-layer wandering, shows a better performance.
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 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.004 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".