clusterBip written by Ignacio Tamarit, María Pereda, and Jose A. Cuesta
This python function is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.
clusterBip is a Python function to cluster bipartite data sets based on the statistical significance of coincidences.
Python version and dependencies: It works both for Python 2.7 and 3.7. You will need to have installed the following libraries: numpy, scipy, matplotlib.
install:
You need to install the libraries at requirements.txt
example of usage:
Firstly, load the function:
from hierarchical_bipartite import hierarchical_bipartite
Then, import some data to play with. Let's analyze the shopping list of 100 people choosing among 20 products to buy. We want to study which products are more likely to be purchased together
datos = pd.read_csv('data/shopping_list.csv', dtype='category', sep=',')
datos = datos.set_index("Labels") #Label column to row names
Let's explore the data
print(datos.head(3))
Ensure your matrix has the entities you want to cluster as rows, and the features as columns:
datos=datos.T
#Get the entities labels
entities_labels=list(datos.index)
Call the function and wait for the magic:
results = hierarchical_bipartite(datos,entities_labels,plot=True,indexes=True, xFontSize=8)
In the dendrogram, you'll see in colors different groups (clusters) of products that are more likely to be purchased together
results[1]
You can obtain an ordered list of the cluster membership (for each entity, the list shows the ID of the cluster it belongs)
results[0]
The normalized susceptibility plot as a function of the p-values of the dendrogram can be shown as:
results[3]
