# Improved Sentiment Detection via Label Transfer from Monolingual to Synthetic Code-Switched Text

arXiv: https://arxiv.org/abs/1906.05725 · HTML: https://mlsanity.com/paper/1906.05725/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Jun 2019.

## Built on by (most cited first, 8 of 8)

- [Language-Driven Image Style Transfer](https://mlsanity.com/paper/2106.00178/) (2021)
- [On the Robustness of Aspect-based Sentiment Analysis: Rethinking Model, Data, and Training](https://mlsanity.com/paper/2304.09563/) (2023)
- [CodeMixBench: Evaluating Code-Mixing Capabilities of LLMs Across 18 Languages](https://mlsanity.com/paper/2507.18791/) (2025)
- [T-STAR: Truthful Style Transfer using AMR Graph as Intermediate Representation](https://mlsanity.com/paper/2212.01667/) (2022)
- [Constructing Code-mixed Universal Dependency Forest for Unbiased Cross-lingual Relation Extraction](https://mlsanity.com/paper/2305.12258/) (2023)
- [Mixed-Distil-BERT: Code-mixed Language Modeling for Bangla, English, and Hindi](https://mlsanity.com/paper/2309.10272/) (2023)
- [Ceasing hate withMoH: Hate Speech Detection in Hindi-English Code-Switched Language](https://mlsanity.com/paper/2110.09393/) (2021)
- [Progressive Sentiment Analysis for Code-Switched Text Data](https://mlsanity.com/paper/2210.14380/) (2022)

## Abstract (arXiv)

Multilingual writers and speakers often alternate between two languages in a single discourse, a practice called "code-switching". Existing sentiment detection methods are usually trained on sentiment-labeled monolingual text. Manually labeled code-switched text, especially involving minority languages, is extremely rare. Consequently, the best monolingual methods perform relatively poorly on code-switched text. We present an effective technique for synthesizing labeled code-switched text from labeled monolingual text, which is more readily available. The idea is to replace carefully selected subtrees of constituency parses of sentences in the resource-rich language with suitable token spans selected from automatic translations to the resource-poor language. By augmenting scarce human-labeled code-switched text with plentiful synthetic code-switched text, we achieve significant improvements in sentiment labeling accuracy (1.5%, 5.11%, 7.20%) for three different language pairs (English-Hindi, English-Spanish and English-Bengali). We also get significant gains for hate speech detection: 4% improvement using only synthetic text and 6% if augmented with real text.