The Tirukural is a classic Tamil Sangam literature authored by Thiruvalluvar. Tirukural comprises of 1130kuratpas. It has been translated into 37 world languages. This necessitates the cross-lingual access of Tirukural on the World Wide Web for which a bilingual dictionary is the primary knowledge base (KB). This KB needs to be constructed. This article puts forth a methodology for automatic construction of a bilingual dictionary for Tirukural in two different languages: Tamil and English. The proposed methodology makes use of the English and Tamil explanatory texts of Tirukural. Explanatory texts in English written by G.U. Pope and that in Tamil by Dr Varadharajan and Dr Solomon Pappaiya are considered in this work. A three-layered model is built using Tirukural and its explanations. Naive Bayes probabilistic learning is used to learn the best mappings between the Tamil and English words. The proposed methodology has been tested with all the 1330 Tamil kuratpas. An efficiency of 70% has been achieved and a performance comparison has been done by using different types of English and Tamil explanatory texts. This method can further be extended to build bilingual dictionaries for other Tamil literatures.
12 August 2025 in Social Network Analysis and Mining
The growing prevalence of hate speech online has amplified acts of discrimination against marginalized populations, with the LGBTQIA+ community being particularly affected. In areas where under-resourced languages such as Malayalam are used, the issue grows more complex because of the absence of localized resources. This research offers an in-depth analysis of the production of counter-speech to address transphobia and homophobia in Malayalam. Our work incorporates both native Malayalam script and Malayalam written in Latin script, addressing the diverse linguistic practices of online users in Kerala. This paper introduces a two-stage pipeline to counter such online abuse. The first stage focuses on dataset creation through a human-in-the-loop process, beginning with 100 seed pairs of hate speech and their corresponding counter-speech manually curated. This set is expanded iteratively using language models culminating in 5,000 validated pairs. In the second stage, we propose a method to generate counter speech in Malayalam that leverages the Retrieval-Augmented Generation framework enhanced by REFINE (Retrieval Evaluation via Fluency, Inversion, and NEarness) for knowledge retrieval and constrained decoding. Evaluation metrics for both dataset quality and model performance demonstrate the effectiveness of this approach in producing diverse, fluent, and target-specific counter-speech. This research provides a foundational resource and scalable strategy for countering hate in low-resource regional languages. GitHub Link: https://github.com/Bharathi-AI-for-Social-Good/CN-Malayalam .
Identifying sarcasm in code-mixed languages presents a pivotal challenge within the domain of natural language processing, given the widespread use of multilingual and multicultural communication on social media platforms. This paper presents findings of the Sarcasm Identification shared task hosted as part of DravidianCodeMix@FIRE-2023. The principal objective of this task was to detect instances of sarcasm within a dataset of code-mixed comments in Tamil-English and Malayalam-English, collected from social media platforms. A total of 11 teams participated in this collaborative effort. This analysis provides a comprehensive assessment of the diverse models employed by the participating teams and the system’s performance was measured in terms of the macro-F1 score.