Package description
Project description
CaGeo: CAnonical Geospatial features
In this repository we want to collect and implement methods to extract
from a set of raw GPS coordinates features that enrich the dataset,
but which, simultaneously also allow dimensionality reduction.
Features:
Point Features
-
Speed: AVG, STD, MIN, MAX
-
Acceleration: AVG, STD, MIN, MAX
-
angle/direction: atan2 == Bearing: angle between the magnetic north and an object ??
Aggregate features
-
Turning Angle: AVG, STD, MIN, MAX. ๐ป๐ถ๐ = |๐๐|/๐ท๐๐ ๐ก๐๐๐๐ Pc is the collection of gps points at which a user changes
his/her heading direction exceeding a certain threshold (Hc), and |๐๐ | represents the number of elements in Pc
-
Traveled Distance: SUM
-
Stop Rate: ๐๐ = |๐๐ |/๐ท๐๐ ๐ก๐๐๐๐ Ps is the collection of point with velocity smaller than a certain threshold
-
Velocity Change Rate: foreach point ๐1. ๐๐ ๐๐ก๐ = |๐2 โ ๐1|/๐1; then ๐๐ถ๐ = |๐๐ฃ|/๐ท๐๐ ๐ก๐๐๐๐ where ๐๐ฃ ={๐๐|๐๐ โ ๐, ๐๐ . ๐๐ ๐๐ก๐ > ๐๐ }
-
FFT?
-
duration of movement?
-
traveled path?
-
displacement?
-
Bearing rate: B_rate(i+1) = (Bi+1 โ Bi)/โt
-
Rate of bearing rate: Br_rate(i+1) = (Brate(i+1) โ Brate(i))/โt
Derivate features
-
sinuosity ?
-
distance from POI
References:
-
A survey and comparison of trajectory classification methods
-
Understanding mobility based on GPS data
-
Revealing the physics of movement: comparing the similarity of movement characteristics of different types of moving objects
-
Predicting Transportation Modes of GPS Trajectories using Feature Engineering and Noise Removal
-
Determination transportation mode on mobile phones
Note
Per le distanze vedere Intelligent Trajectory Classification for Improved Movement Prediction
In "Identifying Different Transportation Modes from Trajectory Data Using Tree-Based Ensamble Classifier": ci sono varie misure globali
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for CaGeo-0.0.4.post1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | a3f6ba69e8240c33418ac3174c109869c823c536543e021ee1e0b85805eb7a26 |
|
MD5 | 475b9b2167258c4b3944d0bed2ef8b54 |
|
BLAKE2b-256 | 91b017af5b2ff880bb944f12d923e4cb62190352095998564170e5b3b1692ae2 |