Improving the Performance of Pakistan's Justice System and its Ranking Using Data Analytics

Abstract

An effective justice system plays an important role in the development of a country. Unfortunately, a large number of lawsuits are pending with judiciary in Pakistan, which indicates the ineffectiveness of the justice system in the country. This contributes to the low rank of the country in the World Justice Project (WJP), which is an international body working for the advancement of the rule of law worldwide. In this paper, we proposed an analytics-based solution to improve the effectiveness of the justice system. The proposed solution consists of two parts. The first part of the solution involves a legal dashboard that presents insights into lawsuits registered with the Supreme Court of Pakistan and provincial courts. This can help to monitor the performance of the justice system and identify areas needing improvement. This can also be helpful in improving the ranking of Pakistan in the WJP. The subjective evaluation of the proposed legal dashboard was done by legal practitioners and law students from renowned local law schools. The majority of the evaluators were of the opinion that the proposed solution can be helpful in improving the effectiveness of the justice system in the country. Past judgments are used as a reference in the current court proceedings. That's why legal practitioners give them due importance and spend a significant amount of time reading them. Usually, after reading the j udgment(s), they find that the judgment(s) did not contain information of their interest, which results in a waste of time and effort. The second part of the proposed solution is a topic modeling technique that can help legal practitioners to find whether the judgment(s) is of their interest or not. We implemented four topic modeling techniques i.e., LDA, LSA, NMF, and HDP using the judgments of the Supreme Court of Pakistan and Islamabad High Court and evaluated their performance on the basis of the coherence CV score. It was revealed that the HDP model is the best technique for extracting topics from court judgments.