Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2025
|
| Online Access: | https://doi.org/10.5281/zenodo.15658091 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866902069044051968 |
|---|---|
| author | Pamulapati.Tejasri, Mr.P.Jayakrishna |
| author_facet | Pamulapati.Tejasri, Mr.P.Jayakrishna |
| contents | <p>The goal of this study is to use decision trees to analyse and predict crimes, since they efficiently handle both numeric and non-numeric (categorical) forms of data. A model is trained using data from old crimes which include details about time, where the offence occurred which crime types and who was involved. In this methodology, tasks are to gather data, clean it, choose the most important features, train models and assess their accuracy. Information for this data gathering comes from police reports and public repositories which is cleaned to ensure it is uniform. The proper features are chosen using metrics for correlation and how important they are. It is developed and optimised using methods called parameter tuning. Some evaluation metrics are precision, recall, F1-score and AUC-ROC. Robustness and the ability to be used in different situations are ensured by cross-validation. It appears that decision trees are able to correctly pinpoint where and when crimes most often happen. The model allows law enforcement to understand data which supports making wise choices</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_15658091 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Crime Data Analysis and Prediction Using Decision Tree Pamulapati.Tejasri, Mr.P.Jayakrishna <p>The goal of this study is to use decision trees to analyse and predict crimes, since they efficiently handle both numeric and non-numeric (categorical) forms of data. A model is trained using data from old crimes which include details about time, where the offence occurred which crime types and who was involved. In this methodology, tasks are to gather data, clean it, choose the most important features, train models and assess their accuracy. Information for this data gathering comes from police reports and public repositories which is cleaned to ensure it is uniform. The proper features are chosen using metrics for correlation and how important they are. It is developed and optimised using methods called parameter tuning. Some evaluation metrics are precision, recall, F1-score and AUC-ROC. Robustness and the ability to be used in different situations are ensured by cross-validation. It appears that decision trees are able to correctly pinpoint where and when crimes most often happen. The model allows law enforcement to understand data which supports making wise choices</p> |
| title | Crime Data Analysis and Prediction Using Decision Tree |
| url | https://doi.org/10.5281/zenodo.15658091 |