VI.应用-Trajectory Data Mining

$textbf{Trajectory Data Mining: An Overview}$

很好的一篇概述，清晰明了地阐述了其框架，涉及内容又十分宽泛。值得细读。

未完成，需要补充。

1. $textbf{Trajectory Data}$：主要分为四个类别
   1. $texttt{Mobility of people}$
   2. $texttt{Mobility of transportation}$
   3. $texttt{Mobility of animals}$
   4. $texttt{Mobility of natural phenomena}$
2. $textbf{Trajectory Data Preprocessing}$
   1. $texttt{Noise Filtering}$
      1. $textit{Mean Filter}$
      2. $textit{Kalman and Particle Filters}$
      3. $textit{Heuristics-Based Outlier Detection}$
   2. $texttt{Stay Point Detection}$
   3. $texttt{Trajectory Compression}$：对轨迹数据进行压缩，以减少计算量
      1. $textit{Distance Metric}$
      2. $textit{Offline Compression}$
      3. $textit{Online Data Reduction}$
      4. $textit{Compression with Semantic Meaning}$
   4. $texttt{Trajectory Segmentation}$：对轨迹数据进行切割
      1. $textit{time interval}$
      2. $textit{shape of a trajectory}$
      3. $textit{semantic meanings}$
   5. $texttt{Map Matching}$：对原始的经纬度数据转化为路网数据
      1. $textit{geometric}$
      2. $textit{topological}$
      3. $textit{probabilis 大专栏  VI.应用-Trajectory Data Miningtic}$
      4. $textit{other advanced techniques}$
3. $textbf{Trajectory Data Management}$
   1. $texttt{Trajectory Indexing and Retrieval}$：没看懂是为了解决什么问题
   2. $texttt{Distance/Similarity of Trajectories}$：了解一下度量方式
4. $textbf{Uncertainty in Trajectory Data}$
   1. $texttt{Reducing Uncertainty from Trajectory Data}$：解决因采样率低，造成数据稀疏，不确定性增大等问题
      1. $textit{Modeling Uncertainty of a Trajectory for Queries}$
      2. $textit{Path Inference from Uncertain Trajectories}$
   2. $texttt{Privacy of Trajectory Data}$：为保护隐私性，需要增大数据的不确定性。
5. $textbf{Trajectory Pattern Mining}$
   1. $texttt{Moving Together Patterns}$
   2. $texttt{Trajectory Clustering}$
   3. $texttt{Mining Sequential Patterns from Trajectories}$
   4. $texttt{Mining Periodical Patterns from Trajectories
      }$
6. $textbf{Trajectory Classification}$：做运动状态分类、交通方式分类等分类任务
7. $textbf{Anomalies Detection From Trajectories}$
   1. $texttt{Detecting Outlier Trajectories}$
   2. $texttt{Identifying Anomalous Events by Trajectories}$
8. $textbf{Transfer Trajectory To Other Representations}$
   1. $texttt{From Trajectory to Graph}$
   2. $texttt{From Trajectory to Matrix}$
   3. $texttt{From Trajectory to Tensor}$

VI.应用-Trajectory Data Mining

原文：https://www.cnblogs.com/lijianming180/p/12262603.html