pssc

KO | EN

gitlite — search

#javascript #ai #typescript #llm #golang #rust #deep-learning #linux #nodejs #java #windows #pytorch #ios #openai

(★ 8)

parallel spatiotemporal spectral clustering algorithm - from my paper

pssc Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# pssc	
[![](https://img.shields.io/badge/maven-v3.2.2-brightgreen.svg)](http://mvnrepository.com/)

Parallel spatiotemporal spectral clustering algorithm.

[中文文档](https://github.com/gyzag/pssc/blob/master/README_CN.md)

The paper ( Parallel Spatiotemporal Spectral Clustering With Massive Trajectory Data ) is published on ISPRS international conference and can be [viewed](http://adsabs.harvard.edu/abs/2017ISPAr42W7.1173G) and [downloaded](https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-2-W7/1173/2017/isprs-archives-XLII-2-W7-1173-2017.pdf) online.

The paper won the conference's best poster.
## Purpose
pssc is an algorithm that aims to mine knowledge from massive trajectory data effectively and efficiently, it expands the time dimension on traditional spectral cluster and speeds up the algorithm by multiple  acceleration solutions.

## Overview
* Use taxi trajectory data as sample data, you can use your own data.
* Use DTW (dynamic time warping) method to calculate spatiotemporal distance.
* Use a  load  balanced  multi-threading method to parallelize the algorithm.
* Use Lanczos method to accelerate the eigenvalue decomposition.

## Data
Demo data is in src/main/resources, which is a small part of real taxi trajectory data. The fields are shown as:

| Field            | Type       | Description               |
| ---------------- |:----------:| -------------------------:|
| CAR_ID           | string     | taxi id                   |
| TIME_STAMP       | string     | time stamp                |
| LONGITUDE        | float      | car longitude             |
| LATITUDE         | float      | car latitude              |
| DIRECTION        | int        | car direction             |
| SPEED            | int        | car speed                 |
| CAR_STATUS       | string     | Carrying passenger or not |

You can use your own data.
## Demo
Demo1 is the cluster result from Wuchang Railway Station to Optics Valley square.
![](https://github.com/gyzag/pssc/raw/master/img/Demo1.png)

Demo2 is the cluster result from Wuchang Railway Station to Wuhan Railway Station.
![](https://github.com/gyzag/pssc/raw/master/img/Demo2.png)

pssc

pssc Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

pssc

pssc Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README