Prediction method of dangerous tendency based on ensemble learning model of behavioral characteristics of inmates

Abstract

The present invention relates to big data processing technology in computer intelligent information processing, and in particular to a dangerous tendency prediction method based on an integrated learning model of behavioral characteristics of prisoners, including: collecting monitoring data of prisoners serving a sentence, preprocessing the data, and treating those without extreme behavior tendencies Mark the data with specific extreme behavior tendency correspondingly; use the marked data as the training set, and use the dispatching ensemble learning algorithm to train on the training set to obtain an ensemble learning model composed of multiple classification models. Then use the integrated learning model to predict and classify the data of unlabeled categories. The beneficial effect of the present invention is that: the prediction method based on the dispatching integrated learning algorithm of the present invention can overcome the defect of poor stability when only one classification model is used by generating a plurality of different classification models and merging their classification results , high classification accuracy, good classification stability, and high early warning accuracy.

Description

technical field [0001] The invention relates to a big data processing technology in computer intelligent information processing, in particular to a dangerous tendency prediction method based on an integrated learning model of behavioral characteristics of inmates. Background technique [0002] With the development of prison informatization, a large number of criminal management business systems including prison administration system, criminal law enforcement system, life sanitation system, family telephone system, interview management system, psychological counseling system, etc. have been built, as well as video surveillance system, Access control system, alarm system, digital power grid, external vehicle and personnel entry and exit management system, emergency command auxiliary decision-making system and a large number of security systems. The construction of these systems provides a large amount of effective basic data for the collection, analysis, and judgment of prison...

AI Technical Summary

Problems solved by technology

[0004] For the existing classification methods, there are also the following deficiencies: (1) When classifying the prisoners, they did not take into account the various types of extreme behavior tendencies that the prisoners may have, such as escape, suicide, violence, destruction, etc.

(2) The existing methods for the classification of inmates rely heavily on artificially designed indicators, the degree of intelligence of the method is not high, and the large amount of data generated by the existing business system is not fully utilized to find abnormal behavior characteristics of inmates

If the prediction performance of the classification model is poor, it will affect the accuracy and stability of the classification in practical applications, so that the ideal classification accuracy and stability cannot be guaranteed.

Images