The task of prison term prediction is to predict the term of penalty based on textual fact description for a certain type of criminal case.Recent advances in deep learning frameworks inspire us to propose a two-step m...The task of prison term prediction is to predict the term of penalty based on textual fact description for a certain type of criminal case.Recent advances in deep learning frameworks inspire us to propose a two-step method to address this problem.To obtain a better understanding and more specific representation of the legal texts,we summarize a judgment model according to relevant law articles and then apply it in the extraction of case feature from judgment documents.By formalizing prison term prediction as a regression problem,we adopt the linear regression model and the neural network model to train the prison term predictor.In experiments,we construct a real-world dataset of theft case judgment documents.Experimental results demonstrate that our method can effectively extract judgment-specific case features from textual fact descriptions.The best performance of the proposed predictor is obtained with a mean absolute error of 3.2087 months,and the accuracy of 72.54%and 90.01%at the error upper bounds of three and six months,respectively.展开更多
This is a replication of Tyler and Bro's study (1992) on the effect of discourse level phenomena on audience perception of comprehensibility. 53 Chinese students of English and 10 native English speakers were take...This is a replication of Tyler and Bro's study (1992) on the effect of discourse level phenomena on audience perception of comprehensibility. 53 Chinese students of English and 10 native English speakers were taken as informants to a questionnaire, in which orders of ideas, discourse miscues and other types of errors (e.g. cohesion and redundant ideas, etc.) were used as variables to see whether they could affect the comprehensibility of texts. Strong resemblances were found between the two groups. Order of ideas (i.e. deductive or inductive) seems not to have affected text comprehensibility much, but the interactive cumulating miscues at the discourse level played an important role in discourse comprehension. As disparities are found between what nonnative speakers do and how they react to what they have done, the paper discusses whether people think the way they write, and if linguistic competence correlates with cognitive ability. The paper suggests that knowing and doing are two aspects of learning; teachers of English, therefore, have to understand the perplexities second language learners face and try to help them write as effectively as possible in the target language.展开更多
基金This work is supported in part by the National Key Research and Development Program of China under grants 2018YFC0830602 and 2016QY03D0501in part by the National Natural Science Foundation of China(NSFC)under grants 61872111,61732022 and 61601146.
文摘The task of prison term prediction is to predict the term of penalty based on textual fact description for a certain type of criminal case.Recent advances in deep learning frameworks inspire us to propose a two-step method to address this problem.To obtain a better understanding and more specific representation of the legal texts,we summarize a judgment model according to relevant law articles and then apply it in the extraction of case feature from judgment documents.By formalizing prison term prediction as a regression problem,we adopt the linear regression model and the neural network model to train the prison term predictor.In experiments,we construct a real-world dataset of theft case judgment documents.Experimental results demonstrate that our method can effectively extract judgment-specific case features from textual fact descriptions.The best performance of the proposed predictor is obtained with a mean absolute error of 3.2087 months,and the accuracy of 72.54%and 90.01%at the error upper bounds of three and six months,respectively.
文摘This is a replication of Tyler and Bro's study (1992) on the effect of discourse level phenomena on audience perception of comprehensibility. 53 Chinese students of English and 10 native English speakers were taken as informants to a questionnaire, in which orders of ideas, discourse miscues and other types of errors (e.g. cohesion and redundant ideas, etc.) were used as variables to see whether they could affect the comprehensibility of texts. Strong resemblances were found between the two groups. Order of ideas (i.e. deductive or inductive) seems not to have affected text comprehensibility much, but the interactive cumulating miscues at the discourse level played an important role in discourse comprehension. As disparities are found between what nonnative speakers do and how they react to what they have done, the paper discusses whether people think the way they write, and if linguistic competence correlates with cognitive ability. The paper suggests that knowing and doing are two aspects of learning; teachers of English, therefore, have to understand the perplexities second language learners face and try to help them write as effectively as possible in the target language.
