High-confidence reconstruction quality and false reduction transient quantitative evaluation method based on cascaded neural network

Abstract

The invention relates to a high-confidence reconstruction quality and false reduction transient quantitative evaluation method based on a cascaded neural network, which is used for carrying out data analysis and extraction on an information database of prisoners and evaluating the reconstruction quality of the prisoners in prisoners. According to the method, a criminal risk assessment scale is used as guidance; the cascade neural network and the migration reconstruction neural network which have a mutual supporting effect are constructed by comprehensively utilizing the full-cycle multi-dimensional related data of the prisoners, and training is performed by using the preprocessed structural data, so that the reconstruction quality evaluation method with high precision and high practicability is obtained. Compared with a traditional model and algorithm without using the neural network, the method provided by the invention is more suitable for multi-dimensional nonlinear original data, the prediction accuracy and efficiency are higher, and it is indicated that the method provided by the invention is effective and practical.

Description

technical field [0001] The invention relates to a cascaded neural network-based high-confidence reconstruction quality and temporary quantitative evaluation method for holiday reduction, belonging to the technical field of neural networks, and in particular to a research method for prison reform quality evaluation and temporary quantitative evaluation of holiday reduction. Background technique [0002] The purpose of reforming inmates is to reintegrate into society and become law-abiding citizens through various means of correction and training. A crime is the measure. [0003] Traditionally, the quality evaluation of inmates' reformation mostly adopts the methods of subjective evaluation or interviews. Traditional retrofit assessment models mostly use risk assessment scales for scoring. This method refers to a list of risk factors, and each item has a standard score form based on the police officer's evaluation experience, and the evaluator scores item by item according t...

AI Technical Summary

Problems solved by technology

However, the current binary logistic regression model is limited to the task of predicting the result of two classifications, and cannot solve the multi-dimensional nonlinear problem of influencing factors, resulting in inaccurate evaluation

The data dimension of the prisoner database is relatively high, and the data types are relatively complex. Binary logistic regression is sensitive to multicollinearity of the independent variables in the model. Two highly correlated independent variables are put into the model at the same time, which may lead to a weaker regression sign of the independent variable. Not as expected, sign is reversed

Images