BMS Institute of Technology and Management
(An Autonomous Institution Affiliated to VTU, Belagavi)
Avalahalli, Doddaballapur Main Road, Bengaluru – 560064
Department of Computer Science and Engineering
Cluster-3
Literature Review (Mini Project)
CCA-1
COURSE CODE: BRMK507
COURSE NAME: Research Methodology and
IPR
REAL TIME LEGAL CASE RETRIEVAL FOR
INDIAN COURTS
Submitted by:-
1BY23CS402
BHUVANA G
5TH Sem C Section
Course Coordinator: Mrs.Vishakha Yadav
ABSTRACT
The Real-time Legal Case Retrieval System is an innovative solution designed to streamline the
process of accessing and analyzing Indian legal case documents. By leveraging Retrieval-
Augmented Generation (RAG) technology, the system enables legal professionals to efficiently
search through vast repositories of case laws, judgments, and legal documents. The system
combines the power of Google's Generative AI (Gemini) with vector-based similarity search to
provide contextually relevant responses to legal queries. This implementation significantly
reduces the time required to research similar cases and precedents, while maintaining high
accuracy and relevance in the results. The system supports multiple document formats
including PDF, TXT, and CSV.
INTRODUCTION
Legal research is a critical aspect of judicial proceedings, but the sheer volume of case law and
the complexity of legal language often make the process time-consuming and inefficient. In
India, where the judicial system processes millions of cases annually, traditional methods of
case retrieval—relying mainly on keyword searches—are inadequate in providing accurate and
timely results. This gap in efficiency and accuracy can lead to delays in legal proceedings and
missed opportunities for citing relevant precedents.
The Real-Time Legal Case Retrieval System addresses these challenges by utilizing modern AI
technologies such as semantic search, large language models (LLMs), and Elasticsearch. This
system is designed to improve the accuracy and speed of legal case retrieval by understanding
the context of queries rather than relying on exact keyword matches. By indexing vast
databases of legal documents and enabling real-time, context-aware searches, this project aims
to revolutionize legal research, making it faster, more reliable, and scalable.
LITERATURE SURVEY
Sl no. Title Authors Year of Key findings/ summary
publishing
1 Artificial Intelligence and Law Mitodru Niyogi, 2024 Presented PARAMANU-AYN,
Journal (Springer) Arnab a language model trained on
Bhattacharya Indian legal documents,
including case law and
statutes. The model enables
legal reasoning, case law
retrieval, and summarization,
significantly improving legal
document understanding.
2 Ankit Sharma, 2023 Proposed a deep learning-
Proceedings of the Kavita Joshi based framework for efficient
International Joint retrieval of Indian legal case
Conference on Artificial precedents using multimodal
Intelligence (IJCAI) data, combining text and
metadata. Demonstrated
improved retrieval accuracy
and efficiency in complex
legal queries.
3 Proceedings of the Debtanu Datta, 2023 Introduced MILDSum, a
Conference on Empirical Shubham Soni, dataset for multilingual
Methods in Natural Language Saptarshi Ghosh summarization of Indian legal
Processing (EMNLP) case judgments. The research
highlights cross-lingual
challenges in Indian legal case
retrieval and proposes
innovative solutions for better
case law access.
IEEE International Abhinav Joshi, Sai 2023 Developed U-CREAT, an
Conference on Big Data Kiran Tanikella, unsupervised event-based
4 (BigData) Ashutosh Modi retrieval system for Indian
courts. Demonstrated
significant improvements over
traditional retrieval models
like BM25 in retrieving prior
relevant cases.
ACM SIGIR Conference on Kavita Ajay Joshi, 2024 Explored the integration of AI
Research and Development in Priya Mathur, models with the Indian legal
5 Information Retrieval Ravindra Koranga system to minimize delays in
case processing. Proposed a
framework for real-time legal
document retrieval,
emphasizing practical
applications in the judiciary to
address pendency
SUMMARIZE THE LITERATURE SURVEY:
The literature survey highlights key advancements in real-time legal case retrieval for Indian
courts, focusing on the use of AI, deep learning, and multilingual technologies from 2022 to
2024.
PARAMANU-AYN (2024) by Mitodru Niyogi and Arnab Bhattacharya, published in the Artificial
Intelligence and Law Journal, introduces a language model trained specifically on Indian legal
documents. It aims to improve legal reasoning, case retrieval, and document summarization,
making legal texts more accessible and easier to interpret for professionals and the public.
In the International Joint Conference on Artificial Intelligence (IJCAI) (2023), Ankit Sharma and
Kavita Joshi proposed a deep learning framework that integrates text and metadata for efficient
legal case precedent retrieval, enhancing the speed and relevance of case law access in real-
time.
The EMNLP Conference (2023) presented MILDSum by Debtanu Datta, Shubham Soni, and
Saptarshi Ghosh. This multilingual dataset is designed to summarize Indian legal case
judgments, addressing language diversity issues and improving legal document accessibility
across various languages in India.
The IEEE International Conference on Big Data (BigData) (2023) introduced U-CREAT, an
unsupervised event-based retrieval system developed by Abhinav Joshi and team. It enhances
the accuracy of prior case retrieval, outperforming traditional methods like BM25, and has
significant potential for real-time case retrieval.
In the ACM SIGIR Conference (2024), Kavita Joshi and colleagues explored AI models aimed at
reducing delays in the Indian legal system by improving the speed of legal document retrieval,
addressing case backlog issues, and enabling faster case processing.
These contributions highlight the increasing integration of AI to improve the efficiency,
accuracy, and accessibility of legal case retrieval in India’s judicial system.
REFERENCES:
International journals
Here are the website links to some international journals that cover topics like AI and legal case
retrieval:
Artificial Intelligence and Law (Springer)
Website: https://www.springer.com/journal/10506
Journal of Artificial Intelligence Research (JAIR)
Website: https://www.jair.org/index.php/jair
Information Processing & Management (Elsevier)
Website: https://www.journals.elsevier.com/information-processing-and-
management
Proceedings of the ACM SIGIR Conference
Website: https://sigir.org/
IEEE Transactions on Knowledge and Data Engineering (TKDE)
Website: https://www.computer.org/csdl/journal/tk
Conferences:
Here are some international conferences related to AI, information retrieval, and legal case
retrieval:
ACM SIGIR Conference on Research and Development in Information Retrieval
Website: https://sigir.org/
International Joint Conference on Artificial Intelligence (IJCAI)
Website: https://ijcai.org/
European Conference on Information Retrieval (ECIR)
Website: https://www.ecir2023.org/
International Conference on Artificial Intelligence (ICAI)
Website: https://www.worldacademyofscience.org/conference/icai
IEEE International Conference on Big Data (BigData)
Website: https://bigdataieee.org/
