MétaCan
Menu
← Back to cohort

Error Analysis for POS Tagging of Hindi-English Code-Mixed Data

2025· article· en· W4414231430 on OpenAlexaff
Xiaoxi Luo

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicNatural Language Processing Techniques
Canadian institutionsUniversity of Waterloo
Fundersnot available
KeywordsSentenceNounTask (project management)HindiSpellingWord (group theory)Error detection and correctionError analysis

Abstract

fetched live from OpenAlex

The phenomenon of code mixing (CM), particularly between Hindi and English, is increasingly prevalent in digital communication, especially on social media. The linguistic features of CM content often contain informal grammar, spelling errors, transliteration, etc., which makes the task of Natural Language Processing (NLP) of such content hard. This paper discusses the implementation of a Part-of-Speech (POS) tagger built using a single-layer Perceptron for CM content on social media and analyses the sources of inaccuracies in tagging. The tagger achieves an accuracy of 84%. The study tries to understand the types of sentence patterns that lead to tagging errors. It shows the correlation of tagging errors with neighbouring word context, sentence length, etc. Certain misclassifications are found to be more common than others, for example, nouns for verbs, which may be related to the difference in the way the two languages place verbs in sentences. POS tagging errors have also been shown to have a cascading effect: one error in a sentence leads to more, and isolated errors are less common. Understanding these error patterns, including common misclassifications, shows how these errors are linked to linguistic features like sentence structure, and highlights areas of improvement for further research.

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 imitation

Not 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.

metaresearch head score (Codex)0.011
metaresearch head score (Gemma)0.054
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.011
Threshold uncertainty score0.059

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0110.054
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0030.003
Science and technology studies0.0010.001
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0010.001

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.

Opus teacher head0.035
GPT teacher head0.339
Teacher spread0.304 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicNatural Language Processing Techniques→French-language works237,207→