GrapHint

Graph-based Time Series Clustering Visualisation Tools

Try our demo

GrapHint is a Python-based web interactive tool to interpet and compare time series clustering on several datasets of the UCR-Archive. It is based on a novel graph-based time series clustering, called $k$-Graph. $k$-Graph is divided into three steps: (i) Graph embedding, (ii) Graph clustering, and (iii) Consensus Clustering. Moreover, we provide a way to select the most interpretable graph for the resulting clustering partition and allow the user to visualize the subsequences contained in the most representative and exclusive nodes. Graphit allows the user the interact with the graphs and identifies the important subsequences that helps creating the clusters. You may find more information here.

Contributors

• Paul Boniol, Inria, ENS, PSL University, CNRS
• Donato Tiano, Università degli Studi di Modena e Reggio Emilia
• Angela Bonifati, Lyon 1 University, IUF, Liris CNRS
• Themis Palpanas, Université Paris Cité, IUF

Usage

Step 1: Clone this repository using git and change into its root directory.

git clone https://github.com/boniolp/graphit.git
cd dsymb-playground/

Step 2: Create and activate a conda environment and install the dependencies.

conda create -n graphit python=3.9
conda activate graphit
pip install -r requirements.txt

Step 3: install Graphviz and pyGraphviz:

• For Mac:

brew install graphviz

• For Linux (Ubuntu):

sudo apt install graphviz

• For Windows:

Stable Windows install packages are listed here

Once Graphviz is installed, you can install pygraphviz as follows:

pip install pygraphviz

Step 4: You can use our tool in two different ways:

• Access online: https://graphit.streamlit.app/
• Run locally (preferable for faster interaction). To do so, run the following command:

streamlit run Hello.py