A Chinese legal intelligent auxiliary discretionary adviser based on GA-BP NNs
| Pages | 1135-1153 |
| Date | 10 December 2018 |
| Published date | 10 December 2018 |
| DOI | https://doi.org/10.1108/EL-03-2017-0056 |
| Author | Ni Zhang,Yi-fei Pu,Suiquan Yang,Jinkang Gao,Zhu Wang,Ji-liu Zhou |
| Subject Matter | Information & knowledge management,Information & communications technology,Internet |
A Chinese legal intelligent
auxiliary discretionary adviser
based on GA-BP NNs
Ni Zhang,Yi-fei Pu and Suiquan Yang
Sichuan University, Chengdy, China
Jinkang Gao
Southwestern University of Finance and Economics, Chengdu, China
Zhu Wang
Sichuan University, Chengdy, China, and
Ji-liu Zhou
Chengdu University of Information Technology, Chengdu, China
Abstract
Purpose –This paper aims to build a legal intelligent auxiliary discretionary system for predicting the
penalty and damage compensation values. After extensively considering current the characteristics of the
current Chinese legal system, a practical legal intelligent auxiliary discretionary system based on genetic
algorithm-backpropagation(GA-BP) neural network(NN) is proposed herein.
Design/methodology/approach –An experiment is designed to analyze cases involving mental
anguish compensation in medical disputes, and a Chinese legal intelligent auxiliary discretionary adviser
system is built based on a GA-BP NN. Because BP neuralnetworks perform well for nonlinear problems and
GAs can improve their ability to find optimal values, and acceleratetheir convergence, a combined GA–BP
algorithm is used. In addition,an ontology is used to reduce the semantic ambiguities and extractthe implied
semanticinformation.
Findings –We confirm that a case-basedlegal intelligent auxiliary discretionaryadviser system based on a
GA-BP NN and ontology techniqueshas good performance in prediction. By predicting the mentalanguish
compensation values, the legal intelligentauxiliary discretionary adviser system can help judges to handle
cases more quickly andordinary people to discover the suggested compensation or penalty.In contrast to BP
NN or SVM, the result seems more closeto the actual compensation rate.
Practical implications –Recently, smart court has been developedin China; the purpose of which is to
build the legal advicesystem for improving judicial justice and reducing differencesin sentencing.A practical
legal advicesystem is an urgent requirement for the judiciary.
Originality/value –This paper presents a study of a case-basedlegal intelligent auxiliary discretionary
adviser system based on a GA-BP NN and ontology techniques. The findings offeradvice to optimize legal
intelligentauxiliary discretionary adviser systemsfor mental anguish compensation in medicaldisputes.
Keywords Artificial intelligence, Information studies, Models, Knowledge-based systems
Paper type Research paper
We would like to thank the Youth Project of the Humanities and Social Sciences of the China Ministry
of Education (No: 15XJC820001); central university basic scientific research project of Sichuan
University (2018skzx-pt193).
Auxiliary
discretionary
adviser
1135
Received11 May 2017
Revised18 October 2017
18December 2017
1 February2018
Accepted21 March 2018
TheElectronic Library
Vol.36 No. 6, 2018
pp. 1135-1153
© Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-03-2017-0056
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
Introduction
Legal auxiliary discretionary adviser (LADA) has received much attention in recent years
because it can reduce sentence deviation and improve judicial justice, facilitate conflict resolution
and also provide judicial data supported for legislation. LADA is a case-based legal expert system
supported by legal big data, such as judicial reports, which provide a good way to discover the
similar cases, helping the judges or other litigation participants to arrive at appropriate penalty or
compensation values. However, the performance of current legal adviser system is far from
satisfactory because the legal discretionary factors are not clear and the technology is immature,
leading to inaccurately simulating legal reasoning or falsely representing the facts.
With the development of artificial intelligence (AI) techniques that use intelligent
machines to simulate human intelligence by studying the essence of intelligence, AI has
been widely used in building legal experts. Since the first legal expert system HYPO was
designed in the 1980s (Ashley, 1992), a number of HYPO-like systems, such as CATO, IBP,
case-based reasoning tool (CABARET) and BankXX, have been designed. Although the
inability of legal expert systems to handle factual uncertainty and uncertainty in applying
the law to particular problems cannot be totally eliminated, the fact that they may facilitate
the creation of legal advice systems makesthem an exciting research topic. The interests of
legal experts has risenand fallen during the past 30 years, ranging from curiosity to faith.In
the background that court information is widely electronic constructed, legal intelligent
auxiliary systemsare atrracting more attention.
Legal intelligent auxiliary systems are designed to improve the efficiency of judges to
help resolve a large number of cases, promote the judicialfairness by limiting the discretion
of judges and aid more people non-legal public by providing them the advantages of
searching similar cases. In particular, in areas, such as spiritual compensation, where the
regulations do not provide clearguidance, case-based analysis is deemed a good method for
reducing deviationsin sentencing and differences in compensation.
China is primarily a rule-orientedcountry depending on written statutes, and case-based
legal systems are seldom studied;however, the construction of wisdom courts from 2016 has
inspired the passion for case-based legal intelligent auxiliary systems. Backpropagation
(BP) neural networks (NNs) are the most popular neural network for their high forecast
precision, favorabledynamic characteristicsand better adaptability and robustness.BP NNs
have two defects: they have low convergence speeds in the learning process, and they get
stuck easily at local optima (Yong etc., 2000). To overcomethe tendency to get stuck in the
local optima of BP NNs, the genetic algorithm (GA), involving to simulate the natural
evolutionary processes to searchthe optimal solution, is adopted to optimize BP algorithms.
A GA-BP NN is proposed in the paper, that has excellent global optimization and rapid
convergence. In addition to eliminating semantic ambiguity of legal cases, ontology
technique is also adoptedin the paper.
This paper aims to build a case-based legal intelligent auxiliary discretionary system
based on GA-BP NN and ontologicaltechniques. The remainder of this paper is organized as
follows. Section 2 reviews the relevant literature. Section 3 describes the proposed legal
expert system and the construction of a framework for judicial cases as well as discusses
how GA-BP NN can be used to build legal expert systems. Section 4 discusses the spiritual
compensation rates suggested by the proposed system. Finally, Section 5 concludes the
paper along with suggestionsfor future work.
Literature review
This section reviews research related to building a case-based legal expert system from
three perspectives:legal expert systems, legal ontologies and BP-GA NNs.
EL
36,6
1136
To continue reading
Request your trial