Metadata-Version: 2.1
Name: crimeanalitictools
Version: 1.1.0
Summary: Statistical methods for studying criminal activity using artificial intelligence (AI)
Home-page: https://github.com/bessonovaleksey/crimeAnaliticTools.git
Author: Aleksey A. Bessonov
Author-email: bestallv@mail.ru
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: Apache Software License
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE.py
Requires-Dist: numpy
Requires-Dist: pandas
Requires-Dist: scipy
Requires-Dist: scikit-learn
Requires-Dist: clustertools-2
Requires-Dist: matplotlib
Requires-Dist: statsmodels

# CrimeAnaliticTools
Statistical methods for studying criminal activity using artificial intelligence (AI)

The package “CrimeanaliticTools” contains functions for a forensic analysis of crimes data, including their exploration analysis and preliminary preparation, some functions for simplifying the use of artificial intelligence methods.

In particular, the package contains the following functions:
Dixon, Grubbs and Irvin test to identify outliers;
computes the area under the curve (AUC), AUC variance and the confidence interval (CI) of AUC;
calculates a cross-tabulation of observed and predicted classes with associated statistics;
variable importance using Garson's and Olden's algorithm, and sensitivity analysis using Lek's profile method for input variables for a Multi-layer Perceptron classifier or regressor;
assessment of the relative importance of features in data for estimator;
search for the best combination of features in data using ensemble methods;
for to draw a Multi-layer Perceptron classifier or regressor in image.
