Open Access

Research on Dynamic Monitoring and Intelligent Early Warning of Community Correctional Recidivism Risk Based on Multidimensional Data Mining

Manna Xie

Manna Xie

Judicial Response College, Guangxi Police CollegeNanning, China
Search for this author on
Sciendo|Google Scholar
Xie, Manna
  
Mar 19, 2025
Research on Dynamic Monitoring and Intelligent Early Warning of Community Correctional Recidivism Risk Based on Multidimensional Data Mining's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Mar 19, 2025
Received: Oct 26, 2024
Accepted: Feb 02, 2025
DOI: https://doi.org/10.2478/amns-2025-0442
Keywords
© 2025 Manna Xie, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Process of multidimensional data mining
Process of multidimensional data mining

Figure 2.

Flowchart of Apriori algorithm
Flowchart of Apriori algorithm

Figure 3.

Flowchart of MApriori algorithm
Flowchart of MApriori algorithm

Figure 4.

DBSCAN schematic diagram
DBSCAN schematic diagram

Figure 5.

The spatial and temporal distribution of the type of crime
The spatial and temporal distribution of the type of crime

Figure 6.

DBSCAN algorithm clustering experimental results
DBSCAN algorithm clustering experimental results

The first multi-dimensional association rule mining results

Serial number Rules Support Confidence Lift
1 (4002-C1)-A1 0.3971 0.9463 2.1641
2 (D1-4002-C1)-A1 0.2879 0.9372 3.0472
3 4002-A1 0.4598 0.9392 1.7959
4 D1-A1 0.5977 0.9475 1.3547
5 (D1-4002)-A1 0.3459 0.9619 2.5961
6 (D1-C1)-A1 0.5130 0.9441 1.7715
7 C1-A1 0.6908 0.9308 1.2436
8 (D1-4002-A1)-C1 0.2898 0.9074 2.5924
9 (D1-4002)-C1 0.3286 0.8353 2.5903
10 D1-C1 0.5399 0.3102 1.5659
11 (D1-A1)-C1 0.4907 0.8178 1.3577
12 (4002-A1)-C1 0.4317 0.8348 1.4991
13 4002-C1 0.4647 0.8543 1.6635

Logistic analysis of community correctional personnel’s recidivism behavior

Influencing factor β SE Wald χ2 OR 95%CI P
Gender
Male 1.02
Female -0.98 0.29 10.32 0.41 0.24~0.71 0.005
Age 22.44 0.000
18-29 years old 1.05
30-39 years old 0.33 0.16 5.04 1.43 1.12~1.85 0.019
40-49 years old 0.31 0.14 4.28 1.14 0.98~1.82 0.025
50-59 years old -4.41 1.12 15.96 0.15 0.02~0.11 0.000
60 years old and above -17.95 3105.68 0.01 0.02 0.963
Domicile 1.05
City -0.38 0.16 7.94 0.72 0.49~0.93 0.007
Countryside
Living condition
Unfixed residence 1.09
Fixed home -3.04 0.35 75.27 0.06 0.02~0.08 0.000
Cultural degree
Primary school and below
Junior high school culture -0.91 0.17 35.14 0.39 0.31~0.56 0.000
High school above -0.82 0.21 2.17 0.78 0.52~1.16 0.144
Occupation
Fixed occupation -4.35 1.06 16.36 0.09 0.01~0.09 0.000
Unemployed man 0.92 0.48 3.85 2.61 0.99~6.38 0.017
Criminal experience
Have drug use experience 1.36 0.95 1.88 0.27 0.05~1.67 0.162
No drug use experience -0.88 0.18 32.95 0.41 0.35~0.56 0.000
Constants 2.61 0.87 9.24 14.24 0.004

Number of different types of crime

Case category 1 2 3 4 5 6 7 8 9 10 11 12 Total
Burglary 75 41 42 72 60 73 66 61 67 68 51 35 711
Braid 2 4 3 3 2 1 1 1 3 0 2 3 25
Drugs 7 6 7 3 1 5 1 4 4 2 0 2 42
Gambling 6 3 5 2 3 4 5 2 7 4 3 1 45
Nuisance class 1 0 2 3 1 1 2 3 4 2 0 1 20
Intentional injury class 0 2 2 7 3 0 2 6 1 4 5 3 35
Traffic accident class 8 0 5 4 11 5 4 6 4 5 0 7 59
Prostitution 2 1 0 3 4 2 1 1 1 1 1 1 18
Blackmail 1 0 0 2 1 3 1 1 1 2 1 1 14
Dangerous driving class 10 11 13 14 22 26 15 10 16 18 16 19 190
Sexual abuse 0 3 1 2 4 1 0 4 2 0 3 2 22
Pick up trouble 3 3 1 4 5 5 3 2 3 3 0 2 34
Scams 51 20 47 45 69 69 87 53 32 47 46 43 609
Other classes 13 5 8 12 9 9 13 11 5 10 3 4 102
Total 179 99 136 176 195 204 201 165 150 166 131 124 1926

The second multi-dimensional association rule mining results

Serial number Rules Support Confidence Lift
1 4002-4002(*) 0.3597 0.6972 1.3024
2 4002(*)-4002 0.3597 0.6038 1.0549
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other
Journal RSS Feed