CrimeDataBD
收藏资源简介:
CrimeDataBD是孟加拉国首个标准犯罪数据集,由阿赫桑科学与技术大学和达卡大学合作创建。该数据集包含2013至2019年间6574起犯罪事件的详细信息,包括时间、地点、天气和人口统计数据。数据收集自孟加拉国最广泛阅读的英文日报《The Daily Star》,并通过手动和自动方式从新闻文章中提取基本特征。此外,还从标准服务提供商获取地理位置和天气数据,以及从孟加拉国国家人口普查报告中收集人口统计信息。数据集通过多种特征工程方法进一步改进,以适应机器学习任务,特别是犯罪预测。该数据集旨在为孟加拉国及其他国家的犯罪预测系统提供基础,帮助执法机构预测和遏制犯罪,优化资源分配。
CrimeDataBD is the first standardized crime dataset in Bangladesh, co-created by Ahsanullah University of Science and Technology and the University of Dhaka. This dataset contains detailed information on 6,574 criminal incidents spanning 2013 to 2019, including temporal, spatial, meteorological and demographic data. The data was collected from The Daily Star, the most widely circulated English-language daily newspaper in Bangladesh, with basic features extracted from news articles via both manual and automated methods. Additionally, geolocation and weather data were obtained from standard service providers, while demographic information was gathered from Bangladesh's national census reports. The dataset has been further refined through multiple feature engineering methods to suit machine learning tasks, particularly crime prediction. This dataset aims to serve as a foundational resource for crime prediction systems in Bangladesh and other countries, assisting law enforcement agencies in predicting and curbing criminal activities and optimizing resource allocation.

- 1Crime Prediction using Machine Learning with a Novel Crime Dataset计算机科学与工程系,阿赫桑科学与技术大学,达卡,孟加拉国 · 2022年



