Legal Question Answering using Ranking SVM and Deep Convolutional Neural Network

TL;DR

This paper explores combining Ranking SVM and CNN for legal information retrieval and question answering, introducing paragraph-level analysis and integrating statistical features for improved accuracy.

Contribution

It proposes a novel approach using paragraph-level features and combined models for legal QA and retrieval, enhancing accuracy over previous methods.

Findings

Improved accuracy in legal question answering.

Effective use of paragraph-level features.

Integration of statistical features enhances CNN performance.

Abstract

This paper presents a study of employing Ranking SVM and Convolutional Neural Network for two missions: legal information retrieval and question answering in the Competition on Legal Information Extraction/Entailment. For the first task, our proposed model used a triple of features (LSI, Manhattan, Jaccard), and is based on paragraph level instead of article level as in previous studies. In fact, each single-paragraph article corresponds to a particular paragraph in a huge multiple-paragraph article. For the legal question answering task, additional statistical features from information retrieval task integrated into Convolutional Neural Network contribute to higher accuracy.