Language identification in multilingual, short and noisy texts using common N-grams
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The problem of Language Identification (LID) has been present in the Natural Language Processing domain for a relatively long period of time. There is a number of approaches based on statistical methods used for this particular task, and lately AI methods with the revival of neural network techniques. Some of the solutions described and implemented in the past show good performance, but texts that were processed were usually clean in terms of grammatical correctness and homogeneity. This paper explores and discusses LID in short and noisy messages written in similar languages, which is a non-trivial task, especially for very related languages. The experimentation methodology in the paper is based on the algorithms such as SVM, Naïve Bayes variants, Random Forest and Logistic Regression. In addition, we explore a novel distance based classification method - Common N-Grams (CNG). Finally, we explored whether Wikipedia as an additional training data source can improve a classifier 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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it