A Primary Research On Machine Learning Based Computer Aided Measure Of Penalty

The Criminal Law our country legislated is the relative uncertain statutory punishment law, and the judge exercise the equitable discretion within the extent for discretionary action of sentencing. However, influenced by many objective and subjective factors, the punishment imparity is existed inevitably. To farthest implement the justice goal that criminal law pursues and get the largest benefit from criminal penalty, the Support Vector Machine (SVM for short), one of the machine learning method that newly emerged in the artificial intelligence theory, is adopted for the application of measurement method research of penalty in this dissertation, and the "Support Vector Machine"measurement model of penalty (SVM penalty measurement model in abbreviation) is presented, which attempted to decrease the imparity in the measurement of penalty through the improvement of penalty measurement method. Then based on the SVM penalty measurement model as the core measurement method of penalty, the machine learning based computer aided measurement of penalty expert system is built and the general frame is described. Finally, take the larceny as the research example, the realization procedures and details of expert system are illustrated. The main part of the dissertation consists of three chapters. In chapter 1, the concept and characteristics of penalty measurement is briefly summarized firstly, which pointed out that as a criminal penalty system, the justice that penalty measurement pursues can be realized only through correct measurement of punishments. Then the requirements that implement correct measurement of penalty is described, and the current existed penalty imparity status is analyzed. Based on these analyses, we can know that, in the current situation, it is emergent and significant to update the penalty measurement method and develop penalty measurement application technique, which can guide the judge to realize the correct measurement of penalty, consequently decrease the imparity of penalty measurement and implement the balance of penalty measurement. Finally, the current research status of penalty measurement methods is summarized, and point out that it is feasible to apply newly emerged machine learning theory to the development of computer aided penalty measurement expert system. Chapter 2 is the key part of this dissertation, in this chapter, the machine learning and support vector machine theory is briefly introduced firstly, and then the feasibility that apply SVM to the development of penalty measurement method is analyzed, followed the model building procedure of SVM penalty measurement model is presented. During the model building process, firstly the expert evaluated samples that are relative correct and can represent the system characteristics are collected. In order to obtain the relative corrected penalty measured samples, the advantages of penalty measurement scheme of UK and USA's legal system are referred to optimize the current penalty measurement scheme that our country adopts, and the correlated penalty measurement scheme theory are analyzed. After the samples are obtained, the act of penalty measurement are extracted and quantified to get the quantity representation of the act, which are then fed as the input to support vector machines for training to get the penalty measurement model. When a new criminal case comes, the act of which are extracted and quantified firstly, then they are sent to the SVM penalty measurement model to obtain the referred penalty measurement. In chapter 3, take the above built SVM penalty measurement model as the core inferential machine, the compute aided penalty measurement expert system is constructed and the general framework is described. Then take the larceny as the research example, the realization procedures and details of expert system are illustrated, which concentrated on the concrete implementation details of SVM penalty measurement model. At the end of the dissertation, the existing problems and further research directions of the expert system are analyzed. In summary, the machine learning theory is adopted for the development of penalty measurement method, and the machine learning based computer aided penalty measurement expert system is built, which realized the crossover between the subjects of criminal law and computer science. However, essentially, the dissertation is researched and written from the viewpoint of criminal law, which put the emphasis on the theory of penalty measurement.