Safest Route Navigation with ML and A* Algorithm for Women Security

The safety of women on different roads while traveling in today's world is very important. So, for women to detect the different crime spots and figure out the safest path can enhance the women safety. In research to detect crime spots we compare the different Machine Learning models: 'Logistic Regression, Grading Boosting Classifier, Support Vector Machine (SVM), KNN, Naïve Bayes’ to detect crime rate and A* Algorithm for calculating a safer path for women. The result of best machine learning model will work as metrics for A* algorithm as crime spots or crime scores, along with separate overall grades for crime based on rates of assault and harassment. These comparative benchmarks to detect crime spot gives accurate result which plays vital role for A* algorithm for finding safest path for women. In order to determine a safer path, our study compares the precision, accuracy and computing efficiency of the different models. The results from the experiments indicate that even Grading Boosting Classifier gives higher accuracy to detect crime spot and generalization on varying environmental conditions than other which has high compute cost to the model and is less applicable for large datasets and A* algorithm for finding path. It also considers how these models can be integrated into real-time navigation systems, focusing on the benefits this is for women in their urban mobility and providing reliable safety assessments.