IIT Kharagpur’s Machine Learning System to Read Legal Documents

Researchers from IIT Kharagpur have recently developed an Artificial Intelligence method to read legal judgments. Machine Learning algorithms play two-fold role by telling which laws are getting violated and by minimizing the legal costs in the judgment process.


The Team

Artificial Intelligence is continually empowering lot of areas by offering the most effective solutions at micro level. IIT Kharagpur researchers have used machine learning (ML) algorithm to develop an automatic method of reading legal documents.
A team of researchers from Department of Computer Science and Engineering applied Machine Learning technique to automate the reading of a legal documents. The team consists of Prof. Saptarshi Ghosh, Department of Computer Science and Engineering from IIT Kharagpur, his students, Paheli Bhattacharya and Shounak Paul, along with researchers from the Tata Research Development and Design Centre, Pune, and Swansea University, United Kingdom.

In the words of Prof. Ghosh, who is leading a team of researchers, “We are trying to build an AI system which can give guidance to the common man about which laws are being violated in a given situation, or if there is merit in taking a particular situation to court, so that legal costs can be minimized”.



Understanding legal judgments itself is tasking. It is important to understand the role of sentences in a legal case judgment. To automate the process of reading the judgement, two deep neural models are used to understand the rhetorical roles of sentences in a legal case judgment. Training deep models has resulted into subsequent tasks of summarization of legal judgments, legal search, case law analysis and related functions. The researchers are using network and text analysis to understand similarity between two legal documents.

Artificial Intelligence is being used in other countries like US, Australia, Japan and Singapore to perform legal research, analyze contracts to check whether they meet required criteria and for reviewing legal documents. After successful application of AI in other fields, India is now set to use it in the legal field. Speaking at an event in Bengaluru, chief Justice of India, Sharad Arvind Bobde said, “We have a possibility of developing Artificial Intelligence for the court system. Only for the purpose of ensuring that the undue delay in justice is prevented”. He also made it clear that there are no plans to use AI in the decision-making process in the courts and AI will never replace human discretion.


Data and the Effort

The team has obtained 50 judgments from the Supreme Court of India. They first segmented these judgements by labeling sentences using multiple human annotators. The project was a good choice by IIT Kharagpur because they have their own law school to suit to the needs of data preparation. Three senior law students from IIT Kharagpur’s Rajiv Gandhi School of Intellectual Property Law were chosen to be human annotators. As next step they performed an extensive analysis of these labels and developed a high quality corpus to train the machine learning algorithm to achieve the objective.

“For a country like India, which uses a common law system that prioritizes the doctrine of legal precedent over statutory law, and where legal documents are often written in an unstructured way, the difference AI can bring is phenomenal,” said Ghosh.

3 Responses to “IIT Kharagpur’s Machine Learning System to Read Legal Documents”

  1. Prerna says:

    Neat post.

  2. Hiral says:

    I like the content you post on your webblog.

  3. Giridhar says:

    Excellent information with extraordinary knowledge.

Leave a Reply

Your email address will not be published.