Title: An efficient approach for storage of court judgements using graph database

Authors: Varsha Mittal; Durgaprasad Gangodkar; Bhaskar Pant

Addresses: Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, India ' Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, India ' Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, India

Abstract: Extracting relevant information from a large dataset has always been challenging and predicting is even more. One such issue comes with arranging large volume of court cases into meaningful data. Text mining approach has its own limitations when it comes to large volume of data having similar type of information. This paper aims at developing an e-dictionary for the courts using a simplified and case wise keyword dictionary creation that could be used to understand the judgements pronounced in earlier similar cases. To overcome the challenge of breaking the text into tokens and get the word count without compromising the speed, a single cluster on Hadoop is considered and finally graph database to store the data. The model is guided by various graph and set theories and is an effort to answer the queries of judges and lawyers in an effective manner by reducing the effort of reading the whole cases and referring only to the final judgement. Use of graph theory provides the logical view of the stored data. To check the robustness of the proposed model precision, recall, and accuracy were calculated and they were found to be 1.0, 0.92 and 92.66%, respectively.

Keywords: Hadoop; graph database; text mining; dictionary; map-reduce; intelligent system; NoSQL databases; natural language processing; NLP; text summarisation; Neo4j; true positive; accuracy; precision; recall.

DOI: 10.1504/IJAIP.2022.123024

International Journal of Advanced Intelligence Paradigms, 2022 Vol.22 No.1/2, pp.214 - 228

Received: 27 Aug 2018
Accepted: 22 Feb 2019
Published online: 23 May 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article