Data Visualisation: Alluvial Diagram vs. Sankey Diagram
This article was published as a part of the Data Science Blogathon.
Most people tend to have a very minimal set of information about Alluvial and Sankey diagrams. The reality is that the two diagrams have superficial similarities to each other. On the flip side, the two diagrams are completely different, underlying superficial assumptions. Let’s discuss these two diagrams briefly in this article!
What is an Alluvial Diagram?
This type of flow diagram represents different forms of changes that made the flow of material, energy, cost, or measurable resource in a certain time-bound. Considering the physical appearance of Alluvial Diagrams and their contribution to the flow, the diagram is named immediately after the alluvial fans.
Note that the fans are formed in a natural way when soil is deposited from streaming water. On the Alluvial Diagram, variables are calibrated on the vertical axis, which is always parallel. Besides, there are values represented with blocks on every axis. The height of each block is meant to represent the height of the streamflow and the size of the container.
The field represents the size of the element within the block and is connected to the stream field directly. Normally, the Alluvial diagram is considered an alternate form of parallel sets although used as categorical variables. It is mainly used to determine trends and time in the long run.
What is a Sankey Diagram?
A Sankey diagram is a form of visualization responsible for depicting a given flow from one set to another. The things that are connected are named nodes, and the connection between these nodes are called links. Sankey Diagrams are mainly used, especially when you want to showcase many-to-many mapping between different domains.
Google Analytics mainly adopts this mechanism to showcase how traffic flows from one page to the other pages of your website. Sankey Diagrams were named after Captain Sankey, who invented the diagram of steam engine efficiency.
In Sankey and Alluvial diagrams, different features make the two confusing. In the two diagrams, nodes are represented by bars that are placed vertically with their height, delivering the idea of volume or count.
The links applied between the nodes consist of curves that have thickness related to the height of the nodes that encodes the count or volume just the same way the node height operates. Also, in all the diagrams, the user needs to highlight a stripe across the diagram. This is to help in identifying how wide the stripe is at different points.
Alluvial diagrams mainly focus on showcasing quantities’ appearance from one state to another throughout different processes. Alternatively, a Sankey Diagram is a streamlined flow chart that can easily visualize quantitative values at every phase of the whole process.
Alluvial and Sankey diagrams have a noteworthy difference considering the consistency of the nodes on the lines. Sankey diagrams have line sets that are uniform in length & analytics while Alluvial diagrams have a lot of variations.
Alluvial Diagrams are mainly applied in multi-dimensional data analysis cases. Emphasis mainly focuses on the frequencies and proportions between different dimensions and how they relate to each other.
Sankey Diagrams are applied in cases that require quantity visualization between various stages of the entire process. Note that a Sankey Diagram can visualize the quantity of the incoming and outgoing parts of the flow. This helps locate the parts with dominant contribution and the places where the quantity gets lost within the process.
The Google charts tools are easy and simple to use when making Sankey diagrams. You can opt to use attractive charts and data tools. Note that the process of making Sankey Diagrams using Google charts is easily customizable.
This mechanism was instigated by Viki Sogn, based in Norway; Highcharts allows its users to make Sankey Diagrams online. You can edit your code as well whenever the need arises.
This is a free tool used for creating Sankey Diagrams by using the drag-and-drop option. You just have to drag and drop your data on the website or import it online.
The DisplayR is an online tool primarily meant for making Alluvial Diagrams. It comes with a bunch of customization options that give users an added advantage when using them. Besides, the tool’s free version does not allow the user to download the final output of the diagram.
This is an automated online tool that requires a web-based application to help you generate Alluvial Diagrams. It gives you a chance to upload your data and generate the final output of your desire. You can easily download the final output or share it using an email.
Alluvial Diagram vs. Sankey Diagram has many things in common. Despite the common features between the two, they are not similar if we review the internal features. They need a deep observation of the physical attributes and their useful capabilities as well. This will help you identify the remarkable similarities and differences between the two.
Alluvial plots are considered a form of Sankey Diagrams that are a notable tool for exploring categorical data. They play a major role in grouping categorical roles into the flows that can be easily tracked and analyzed by the diagram.
It is no secret that Sankey Diagrams are applied in different scenarios daily. The diagrams are mainly applied in the visualization of material, energy flow, and cost. Alternatively, they showcase energy or mass flows in the form of arrows proportional to the size of the flow.
● Download the dataset and the questionnaire responses from Dropbox.
● Click and open the RAW.
● Drag your preferred data into the box.
● Choose the Alluvial Diagram from the list.
● Opt for various dimensions.
● Increase the width of the visualization.
● Amped the visualization.
The alluvial diagram is a type of flow chart that represents changes that are made in the flow of material, energy, cost, or measurable resource over time. In that sense, it helps identify patterns and trends. It is named after the alluvial fan, a flow of water extending over a surface that becomes flat, so it refers to both its shape and flow.
The media shown in this article is not owned by Analytics Vidhya and is used at the Author’s discretion.