IHLawRecommender: Deep Semantic Modelling for IPC Case

Recommendation with Legal Domain Constraints

Gautham Praveen Ramalingam1,* Dharini Devi Ramalingam2 A. Afrin Farhan3

1 Research Scholar, Syed Ammal Engineering College, Ramanathapuram – 623502, Tamilnadu, India

2 Graduate, Government Law College, Madurai – 625020, Tamilnadu, India

3 Student, Syed Ammal Engineering College, Ramanathapuram – 623502, Tamilnadu, India

Received: January 17, 2026 Revised: February 25, 2026 Accepted: March 20, 2026 ⋆ Corresponding author

ABSTRACT

Efficient retrieval of relevant legal cases is critical for judicial decision-making, particularly for high-severity

crimes where timely reference to precedents can influence outcomes. Our work presents IHLawRecommender,

i.e., Intelligent Hybrid Law Recommender, a hybrid framework for recommending Indian Penal Code (IPC) cases

based on textual descriptions provided by users. The system operates through a multi-stage workflow: first, case

descriptions are normalized to remove inconsistencies and embedded into semantic vectors using a Bi-directional

Long Short-Term Memory (BiLSTM) network. These embeddings are compared with the user query to measure

semantic similarity. In parallel, an IPC-specific keyword map evaluates the relevance of each case, while legal-aware

filters distinguish between sexual and non-sexual violent crimes to ensure contextually appropriate recommendations.

The outputs from these stages are integrated using a weighted payoff function that considers semantic similarity,

keyword relevance, and crime severity to produce a ranked list of top-k cases. The system also provides interpretable

visualizations, including heatmaps that illustrate correlations between similarity, keyword score, severity, and payoff.

Evaluation on a curated IPC dataset demonstrates that IHLawRecommender consistently prioritizes legally critical

cases, reduces irrelevant matches, and offers a practical, workflow-driven tool for legal professionals to efficiently

navigate case law while maintaining adherence to judicial priorities.

Keywords: Legal Case Recommendation Bi-directional Long Short-Term Memory (BiLSTM) Semantic Embedding

Keyword-Based Scoring Hybrid Recommendation System

1. INTRODUCTION

In contemporary judicial systems, managing the growing

volume and complexity of legal case records has become a

significant challenge. Accurate and timely legal decisions

require systems that can reason under uncertainty, understand

contextual nuances, and prioritize cases based on offense

severity. Traditional rule-based or deterministic legal support

tools, while effective in static scenarios, often lack the flexibility

to accommodate real-world legal ambiguity and variability

across cases. Rule-based methods provide a mechanism to

assess relevance by mapping query terms to offense-specific

keywords. They have been employed in numerous applications

such as legal information retrieval, precedent analysis,

and case ranking, where domain-specific terminology can

replace simplistic exact-match searches [1, 2]. However, rulebased

approaches alone are limited in capturing semantic similarity,

contextual relationships, and long-term dependencies

across multiple case descriptions. In contrast, semantic similarity

measures, derived from statistical and embedding-based