相关实验视频
Updated: Jan 20, 2026

Identifying Amino Acid Overproducers Using Rare-Codon-Rich Markers
Published on: June 24, 2019
超越规范:使用无监督学习识别罕见和高风险的行人撞车模式
Zeinab Bayati1, Asad J Khattak1
1Department of Civil and Environmental Engineering, The University of Tennessee, Knoxville, TN, United States.
步行者安全需要专注于高风险的"边缘案例"碰撞. 一个新的框架确定了这些罕见的,严重的事件,通常被传统方法忽视,使得有针对性的安全干预成为可能.
科学领域:
- 运输安全运输安全
- 交通事故分析 交通事故分析
- 公共卫生 公共卫生
背景情况:
- 步行者安全是一个关键问题,尽管现有改进,死亡人数仍在增加.
- 传统和自动化车辆安全的进步需要专注于最危险的碰撞场景.
研究的目的:
- 引入和验证复合无监督边缘病例检测框架,用于识别高风险行人撞车事故.
- 分析不同类型事故的特征和严重程度,区分常见模式与罕见的复杂事件.
主要方法:
- 利用统一的多重近似和投影 (UMAP) 来减少维度和基于层次密度的噪音应用的空间集群 (HDBSCAN).
- 开发了一个基于集群成员的不确定性和与典型撞车模式的距离的复合评分系统.
- 使用行人和自行车撞车分析工具 (PBCAT) 将框架应用于北卡罗来纳州警方报告的10,108起行人撞车事故.
主要成果:
- 该框架分类崩到核心,中等边缘和强边缘类别.
- 强边缘碰撞的严重程度明显更高,其中36.6%导致致命伤害,而核心组的8.1%.
- 高风险撞车经常与农村地区,照明不良,非十字路口位置和不寻常的行人行为有关.
结论:
- 开发的边缘案例框架有效地检测罕见的,严重的行人撞车,传统方法可能会忽视.
- 撞车严重程度受到建筑环境和撞车类型的重大影响,突出了需要特定环境的安全策略的需要.
- 通过识别和解决这些高风险边缘案例情景,可以加强有针对性的安全工作.
相关概念视频
A Minority of One: Conformity to Group Norms
It is obvious that we are influenced by those around us, but in the early to mid 1900's, psychologists began to study how potent social influence can be on our thoughts and behaviors. Motivated in part by attempts to explain the behaviors of Nazi soldiers in World War II, one topic of considerable interest at the time in psychology was conformity, the phenomenon in which people match their attitudes, behaviors, or beliefs to...
10:41Identifying Amino Acid Overproducers Using Rare-Codon-Rich Markers
16:14Trajectory Data Analyses for Pedestrian Space-time Activity Study
10:52Simulation of Human-induced Vibrations Based on the Characterized In-field Pedestrian Behavior
06:38Using a Virtual Reality Walking Simulator to Investigate Pedestrian Behavior
08:51Application of Unsupervised Multi-Omic Factor Analysis to Uncover Patterns of Variation and Molecular Processes Linked to Cardiovascular Disease
