Work

AI-Based Detection of Caste and Religion-Based Inequality with Legal Awareness System in India

Year: 2026

Type: article

Abstract: This paper presents a Streamlit-based web application for detecting caste and religion-based discrimination in textual content using a rule-based scoring approach. The system integrates text preprocessing, severity classification, SQLite-based analytics, and legal awareness features aligned with Indian constitutional provisions such as Article 15 and IPC Section 153A. The platform provides real-time feedback, dashboard visualization, and exportable reports, making it suitable for low-resource environments. The system achieved an overall F1-score of 90.8% and demonstrates high interpretability and efficiency for ethical AI-based content moderation in India.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.19200130

Authors: Ishwariya, B, Arthi, R, Muthumeena, S, Pritika, BL, Vasanthi, R

Language: English

Topic: Hate Speech and Cyberbullying Detection

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDGs: Peace, justice, and strong institutions, Reduced inequalities

Open Access status: green

OpenAlex ID: W7140189893

FWCI:

Cites:

Cited by:

Related to: