Crime is one of the most critical problems in urban centers, especially in large cities. In this sense, technological solutions are needed to provide security to citizens, contributing to the reduction of crime rates. The present work proposes the UFollower model (Ubiquitous Follower) which meets this scenario. The scientific contribution of this work consists of the use of Context Histories and User Profiles for data analysis focused on ubiquitous security. The use of prediction mechanisms and historical contexts allowed to reach up to 70.35% of inference rate for a particular crime (domestic violence). The comparison with the related works indicates that UFollower is the only proposal that presents the issue of public security with context histories and user profile management. The evaluation was conducted through scenarios, allowing to evaluate the related hypotheses. In this sense, a context simulator was built where twenty objects interacted to allow the evolution. Among them, there are people, vehicles, cameras, and wearables.

中文翻译：

---

UFollower：基于无处不在的安全和监控的智慧城市模型

犯罪是城市中心，尤其是大城市中最严重的问题之一。从这个意义上说，需要技术解决方案来为公民提供安全保障，从而有助于降低犯罪率。目前的工作提出了满足这种情况的 UFollower 模型（Ubiquitous Follower）。这项工作的科学贡献包括使用上下文历史和用户配置文件进行数据分析，重点是无处不在的安全性。预测机制和历史背景的使用允许对特定犯罪（家庭暴力）达到高达 70.35% 的推理率。与相关工作的比较表明，UFollower 是唯一提出具有上下文历史和用户配置文件管理的公共安全问题的提案。评估是通过情景进行的，允许评估相关假设。从这个意义上说，建立了一个上下文模拟器，其中二十个对象相互作用以允许进化。其中，有人、车、相机、可穿戴设备。