A Markov Logic Networks Based Method to Predict Judicial Decisions of Divorce Cases

Prediction of the judicial decision of a case is a research issue of artificial intelligence in legal domain. Existing studies mainly focus on criminal cases and aim at charge prediction, moreover the results of these models are usually hard to interpret. In this paper we propose a Markov logic networks based method for this problem. We firstly describe and extract the semantic of legal factors in a formal way; then we build and train a Markov logic networks for the prediction. The experimental results of prediction for divorce cases show that, our method is insusceptible to different expression styles, at the same time its prediction outcomes are interpretable.