What is a network visualization tool?

A network visualization tool allows you to visualize nodes and edges as a network. What do we mean by nodes and edges?

Well a node is any type of entity. In the context of organizations and organizational network analysis nodes are generally people or employees of the organization. But Polinode's network analysis tool is very flexible and capable of analyzing literally any networked or

And edges? Well, edges are just connections between the nodes in the network. So, in the context of organizations, those connections are most often working relationships between two individuals. But, again, edges can depict literally any type of relationship at all.

How do you import network data?

Polinode's network tool allows you to import network data in three different ways:

Directly importing the data from an Excel file or a GEXF or JSON file.

ou can import literally any networked or connected data into Polinode and the most common way for people to import data is via an Excel file. We also support more advanced formats such as GEXF or JSON.

Running a relationship-based survey

Polinode contains a separate survey tool designed to collect relationship data by asking people questions such as who do you work with often or who do you go to for advice? It's possible to then simply click a button and the resulting networks will be available in Polinode's networks tool for network analysis and visualization.

By running a passive data query

Polinode's passive data integrations allow you to collect metadata from communication data such emails, calendars and enterprise social networks such as Teams and Slack. You can then simply run a query against this data and the result will appear in Polinode's interactive networks tool.

Features

What are the key features you should look for in a network visualization and mapping tool?

So, all network visualization tools allow you to import connected data and then visualize it as a graph according to a layout. But what are the key features that you should look out for beyond these basics?

Flexibility and ease of use

Polinode's network analysis tool is highly flexible - it allows you to import data that represents any set of nodes and edges and to also attach any attributes for those nodes and edges (which you can think of as metadata).

It's also easy to use - it's possible to import data from Excel in a matter of minutes but it also supports other data formats such as GEXF, which allows you to bring existing data from other tools that you might already use.

Network maps can be shared

Most of the time when you are creating network diagrams, you want to be able to then share those diagrams with others. With Polinode this is not only possible, we make it easy. You can share your interactive networks with others by either giving them access to those networks as a user or, if you choose to make your network public, by just sending them a link to that network.

Capture multiple views for analysis

Typically you will want to slice and dice the network data in multiple different ways. For example, filtering in order to analyze a sub-network and then performing a number of operations on that sub-network. Polinode allows you to do this - each network can have up to 50 different and distinct views saved and applied to it.

Security of the software

Network data, particularly in the context of organizations, can be sensitive data. It's important to know that your private data is protected and secured. Polinode complies with enterprise security best practices and is SOC 2 certified. We take the protection of your data incredibly seriously.

Scales to large networks

Not every network is large but, where you do have a large network, you will want to make sure that the graph visualization software that you use is able to scale to the number of nodes and edges that you have. This is both in terms of the actual visualization as well as the metrics (such as betweenness centrality) that the software allows you to calculate.

Polinode is designed to scale to large networks and can easily handle up to 50,000 nodes and up to 250,000 edges.

Intuitive UX for interactivity

The final important consideration when selecting graph visualization software is the user experience - is the interface intuitive for users to use? Networks can be complex but you don't want software that is overly difficult to use.

Polinode emphasizes the balance between ease of use and powerful network insights and data visualization. You can find a short video demonstrating this ease of use here.

How do I share network visualizations?

As browser-based SaaS software, one of the advantages of Polinode is that it is easy to share your network visualizations (and saved views). If the network is a public network all you have to do is copy the url to share it with someone else - they don't need a Polinode account to access that network (or saved view).

On the other hand, if your network is a private network, you can simply add them as a user to the network once they have their own Polinode account. It's even possible to give them read-only access and to lock down who is able to add or remove other users from accessing the network.

What network layouts are supported?

The most commonly used network layout in Polinode is the Force Directed layout. This layout simulates both attractive and repulsive forces such that the net result (once an equilibrium is reached) is that nodes that are relatively closely connected to each other are positioned relatively close to each other in space.

The Force Directed layout that Polinode uses is based on the Force Atlas 2 layout algorithm - arguably the most effective (yet efficient) algorithm available.

‍

We also support a number of other layout algorithms, including the lens layout, distribute nodes, plot nodes and a tree layout that allows you to create an org-chart layout using the formal reporting structure of an organization (if you are using Polinode with organization data).

‍

What network metrics can I calculate?

Polinode supports over 30 different metrics. These metrics fall into three categories - node metrics, edge metrics and overall network metrics. Details of all of these metrics can be found in our written documentation.

Our node metrics include basic metrics such as In Degree, Out Degree and Total Degree as well as more advanced metrics such as Betweenness Centrality and Page Rank. Edge metrics allow you to easily calculate edge Betweenness as well as whether an edge is reciprocated or not.

And, finally, overall network metrics include Average Shortest Path, Network Density and Diameter to name a few.

Are large networks handled well?

We pride ourselves on our ability to handle large networks - at least up to 50,000 nodes and 250,000 edges and often beyond that. Polinode uses your GPU to visualize the network and we also have very scalable server-side metrics to ensure that we can calculate computationally demanding metrics such as Betweenness centrality for large networks.

Example output of finding influencers via organizational network analysis.

What other network visualization software can I compare Polinode to?

We are often asked how Polinode compares to other graph visualization tools and have prepared a summary comparing Polinode to a few of the most commonly cited alternative tools below.

Gephi

Gephi is an open-source tool that is capable of handling quite large networks. It's written in Java and is a desktop tool rather than being a SaaS tool like Polinode. So, it's not possible to easily share your networks or insights with others. It can also be tricky to install given the Java dependency and does not come with support or assistance.

UciNet

UciNET is a tool that is primarily used by academics and dates back to the early 1980s. UciNET is Windows-only software and is not a SaaS tool which means it is hard to share your networks and insights with others. The network visualization and some of the metrics in UciNET do not scale well to medium to larger size networks.

Pajek

One of Pajek's major advantages is that it handles large networks very well (particularly Pajek XXL and Pajek 3XL). However, it is still a desktop application that means it is difficult to share and collaborate with others. We also often hear that the user interface of Pajek can be quite unintuitive.

NetworkX

NetworkX is a Python library and is open source. It has the advantages that come with being a Python library - it's quite flexible. However, coding knowledge is required to use it and lacks the interactivity of a SaaS solution like Polinode

iGraph

iGraph is also open source but is an R package. Similar to NetworkX, it can be both powerful and flexible but requires a good knowledge of R to use it and it's not really possible to share results and insights with others like it is in Polinode. We do find that some of our users use Polinode in conjunction with other tools like NetworkX and iGraph - exporting and importing data between Polinode and other tools

How can I get started with network analysis?

The easiest way to get started is to sign up for a free Polinode account. You may also want to request a demo so that we can understand your particular use case better and give you some advice and pointers on getting started. And, finally, we always suggest watching our tutorial videos.

Network Visualization Tool