# Detecting Anchors' Opinion in Hinghlish News Delivery

arXiv: https://arxiv.org/abs/2204.02155 · HTML: https://mlsanity.com/paper/2204.02155/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 16 references that are in the map.
- First posted on arXiv 5 Apr 2022.

## Builds on (most cited first, 8 of 8)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [AI4Bharat-IndicNLP Corpus: Monolingual Corpora and Word Embeddings for Indic Languages](https://mlsanity.com/paper/2005.00085/) (2020)
- [Towards Sub-Word Level Compositions for Sentiment Analysis of Hindi-English Code Mixed Text](https://mlsanity.com/paper/1611.00472/) (2016)
- [From English to Code-Switching: Transfer Learning with Strong Morphological Clues](https://mlsanity.com/paper/1909.05158/) (2019)
- [Towards Code-switched Classification Exploiting Constituent Language Resources](https://mlsanity.com/paper/2011.01913/) (2020)
- [HIT: A Hierarchically Fused Deep Attention Network for Robust Code-mixed Language Representation](https://mlsanity.com/paper/2105.14600/) (2021)

## Abstract (arXiv)

Humans like to express their opinions and crave the opinions of others. Mining and detecting opinions from various sources are beneficial to individuals, organisations, and even governments. One such organisation is news media, where a general norm is not to showcase opinions from their side. Anchors are the face of the digital media, and it is required for them not to be opinionated. However, at times, they diverge from the accepted norm and insert their opinions into otherwise straightforward news reports, either purposefully or unintentionally. This is primarily seen in debates as it requires the anchors to be spontaneous, thus making them vulnerable to add their opinions. The consequence of such mishappening might lead to biased news or even supporting a certain agenda at the worst. To this end, we propose a novel task of anchors' opinion detection in debates. We curate code-mixed news debates and develop the ODIN dataset. A total of 2054 anchors' utterances in the dataset are marked as opinionated or non-opinionated. Lastly, we propose DetONADe, an interactive attention-based framework for classifying anchors' utterances and obtain the best weighted-F1 score of 0.703. A thorough analysis and evaluation show many interesting patterns in the dataset and predictions.