The Clustering of Water Consumption with SOM (Self-Organizing Maps) Algorithm in Python

<p>Hello everyone!</p> <p>Today, I would like to share one of my college projects, namely the clustering of water consumption by the Jetis Kulon society. Jetis Kulon is a place, located in GunungKidul, DI Yogyakarta, Indonesia. In this article, we will use the SOM Algorithm to make a cluster model.</p> <p>Before we start training our data in python code, it is a good idea to recall our knowledge regarding the SOM algorithm first.</p> <h1><strong>Introduction of SOM Algorithm</strong></h1> <p>SOM is an unsupervised learning algorithm that was proposed by Kohonen in 1981. Unlike other artificial neural networks that usually use error-function evaluation, SOM uses competitive learning. In this network, a layer containing neurons will arrange itself based on certain input values ​​in a group known as a cluster. During the self-arrangement process, the cluster that has the weight vector that best matches the input pattern (has the closest distance) will be selected as the winner. The winning neuron and its neighbours will improve their weights.</p> <h1>Understanding the Data</h1> <p>In this article we will use water consumption by the Jetis Kulon Society dataset. You can download the dataset by this&nbsp;<a href="https://drive.google.com/file/d/1EMltgpfZNCOS75vsUkavNnaJ9Te1-GQv/view?usp=drive_link" rel="noopener ugc nofollow" target="_blank">link</a>. The dataset was obtained from interviews with 42 residents of Jetis Kulon. There are 7 columns and 42 rows in this dataset. The detailed information about the dataset is provided in the table below.</p> <p><a href="https://medium.com/@vivinsativa1/the-clustering-of-water-consumption-with-som-self-organizing-maps-algorithm-in-python-59b26f05c457"><strong>Click Here</strong></a></p>