eren-ck/st_optics

ST-OPTICS: Spatio-temporal clustering based on OPTICS

ST-OPTICS

Simple and effective tool for spatial-temporal clustering

st_optics is an open-source software package for the spatial-temporal clustering of movement data:

• Implemnted using numpy and sklearn
• Enables to also scale to memory - with splitting the data into frames
• Usage: can view a demo of common features in this this Jupyter Notebook.

Installation

The easiest way to install st_optics is by using pip :

pip install st_optics

How to use

from st_optics import ST_OPTICS

st_optics = ST_OPTICS(xi = 0.4, eps2 = 10, min_samples = 5)
st_optics.fit(data)

Description

A package to perform the ST OPTICS clustering. For more details please see the following papers:

• Ankerst, M., Breunig, M. M., Kriegel, H. P., & Sander, J. (1999). OPTICS: ordering points to identify the clustering structure. ACM Sigmod record, 28(2), 49-60.

License

This package was developed by Eren Cakmak from the Data Analysis and Visualization Group and the Department of Collective Behaviour at the University Konstanz funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's ExcellenceStrategy – EXC 2117 – 422037984“

Released under MIT License. See the LICENSE file for details.

st_optics's Issues

Details Regarding ST OPTICS Clustering

Sir, I have come across your python code for ST_OPTICS and ST_DBSCAN. The codes were extremely useful and informative on the technical side but I have some doubts regarding the output dimension and labeling details. I have an input Geotiff file of 3 dimensions (space and time), Do i expect to get a 3 dimensional output or 2D?. How the labeling is done?, what differentiates the labels from each other?, is there any way to identify. Where shall i get more information about the spatio-temporal OPTICS and DBSCAN or visualise how the data is processed?. Also how to get optimum values for parameters like eps, max_eps, min points etc. I am relatively new to the topic so please Ignore any mistakes in the questions. Hope that you respond positively.