9.1 sgnc demo

20210322 The pre-built demonstration highlights the capabilities of the package. It is self explanatory.

ml demo sgnc

A sample of the interactions:

====================================
StellarGraph for Node Classification
====================================

Welcome to a demonstration of node classification in a graph knowledge
structure. StellarGraph is used to represent the graph. The sample
dataset is a well known public network dataset known as Cora. It is
available from linqs.soe.ucsc.edu/data and consists of nodes which are
academic publications and edges that link citations. The nodes have
been classified  into seven subject areas as we will see below.

A Graph Convolution Network (GCN) is used to build a classification
model to predict the subject area of a publication based on the graph
structure. This neural network model uses a graph convolution layer
which uses the graph adjacency matrix to learn about a publication's
citations.

This demonstration will prepare the dataset, create the GCN layers,
and then train a model and evaluate its performance.

Press Enter to continue: 

===================
Dataset Description
===================

The dataset, available through the StellarGraph package itself, has
been attached and is ready to be loaded into the StellaGraph data
structures.

The Cora dataset consists of 2708 scientific publications classified
into one of seven classes. The citation network consists of 5429
links. Each publication in the dataset is described by a 0/1-valued
word vector indicating the absence/presence of the corresponding word
from the dictionary. The dictionary consists of 1433 unique words.

...

======================
Machine Learning Model
======================

A model is now being built. This is the Graph Convolution Network
model. For a small dataset it takes a few seconds.

Using GCN (local pooling) filters...

...


Your donation will support ongoing availability and give you access to the PDF version of this book. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Popular open source software includes rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2021 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0.