Downloads: 1
United States | Information Technology | Volume 14 Issue 11, November 2025 | Pages: 1978 - 1982
AI-Based Methodology for Legal Document Analysis: Increasing Access to Justice for Vulnerable Communities
Abstract: The global access to justice gap is a persistent and acute social problem that disproportionately affects vulnerable groups who systematically face barriers to obtaining legal assistance. This study assesses the effectiveness and limits of an artificial intelligence (AI) based hybrid methodology aimed at the preventive identification of legal risks in standard-form documents. The aim of the research is to determine whether such a tool can become a scalable mechanism for expanding access to justice. The proposed approach combines natural language processing (NLP) methods for semantic text analysis with a curated expert knowledge base on legal risks, which ensures both high accuracy and interpretability of the outputs. Based on the analysis of more than 12000 documents (lease agreements, employment contracts, consumer agreements), the system proved effective in detecting potentially unfavorable and onerous terms, demonstrating an average F1 measure of 0.89. It is concluded that the described technology can shift legal aid practice from a reactive to a proactive model, providing citizens with tools for independent risk assessment prior to entering into legally significant agreements. The results are of interest to researchers in legal technologies, practitioners of the legal aid system, and policymakers in the sphere of justice.
Keywords: Access to justice, vulnerable populations, legal technologies (Legal Tech), natural language processing (NLP), document analysis, risk management, explainable AI (XAI), algorithmic bias, preventive law, computational jurisprudence
How to Cite?: Ferents Filip, "AI-Based Methodology for Legal Document Analysis: Increasing Access to Justice for Vulnerable Communities", Volume 14 Issue 11, November 2025, International Journal of Science and Research (IJSR), Pages: 1978-1982, https://www.ijsr.net/getabstract.php?paperid=SR251104085601, DOI: https://dx.doi.org/10.21275/SR251104085601