Role of Artificial Intelligence in Detection of Hateful Speech for Hinglish Data on Social Media

Abstract

Social networking platforms provide a conduit to disseminate our ideas, views and thoughts and proliferate information. This has led to the amalgamation of English with natively spoken languages. Prevalence of Hindi-English code-mixed data (Hinglish) is on the rise with most of the urban population all over the world. Hate speech detection algorithms deployed by most social networking platforms are unable to filter out offensive and abusive content posted in these code-mixed languages. Thus, the worldwide hate speech detection rate of around 44% drops even more considering the content in Indian colloquial languages and slangs. In this paper, we propose a methodology for efficient detection of unstructured code-mix Hinglish language. Fine-tuning based approaches for Hindi-English code-mixed language are employed by utilizing contextual based embeddings such as ELMo (Embeddings for Language Models), FLAIR, and transformer-based BERT (Bidirectional Encoder Representations from Transformers). Our proposed approach is compared against the pre-existing methods and results are compared for various datasets. Our model outperforms the other methods and frameworks.

Explainable Misinformation Detection Across Multiple Social Media Platforms

“…The information on social media is usually easygoing and casual, leading to a decrease in the ability of a language model to comprehend the corpus; consequently, performing broad preprocessing on the information became necessary [72]. Fig 1 depicts the diagrammatic flow for preprocessing.…”

Multimodal Hinglish Tweet Dataset for Deep Pragmatic Analysis

“…Still, Hinglish sentiment analysis might be hard because it uses two different languages and rules and frameworks that aren't usually used in language [12].Text pragmatic analysis, on the other hand, looks at how language is used to get something done in a certain setting. To do this, you have to look at a text's meaning beyond its formal meaning [13], taking into account the speaker's intentions, the audience, and the social and cultural setting in which the work is created [14].…”

“…It is difficult to construct accurate algorithms for sentiment and pragmatic analysis in the absence of appropriate data such as for low-resourced language [17]. Beside this, the complexity of Hinglish, which mixes two distinct grammatical systems and vocabularies, makes it challenging to develop suitable models for machine learning techniques [13]. Lack of standardisation is yet another big issue [18][19][20].…”