Graph-o-mania: The Flowering of a New Visual Paradigm in Business Analytics

Big Data Evangelist, IBM

Graph analysis is all the rage these days, and not just among data scientists. It's not a new discipline, but, to the larger business and consumer public over the past few years, it has seemingly come out of nowhere.

Graph analysis is, at heart, a mathematical approach for mapping complex relationships among networks of nodes. Graph modeling is an established branch of statistical modeling that focuses on mining, mapping, visualizing, and exploring connections, interactions, and affinities. What distinguishes graph analysis is a focus on "graphs," which are abstract networks of relationships (known as "links") among "nodes" (which may be individuals, groups, companies, products, systems, objects, concepts, words, and other entities). In addition to applications in social and semantic applications, graph analysis has well-established uses in scientific, engineering, and other domains.

Of the technology's many uses, social graph analysis is most popular, thriving on the gusher of customer intelligence flowing from online communities of all shapes and sizes.  In addition to customer profiles and other contextual data, modelers may incorporate a huge range of behavioral information into social graph models. The behavioral data sources might include Facebook status updates, tweets, portal clickstreams, geospatial coordinates, transaction records, interest profiles, call detail records, and usage logs. Social graphs may also incorporate diverse streams of big data--structured and unstructured, user- and machine-generated, etc.--that issue from social media as well as from B2C communities, B2B supply chains, and enterprise applications.

The recent mania for all things "graph" stems in part from increased use of the word in social-media contexts. Most noteworthy is Facebook's recent rollout of the "graph search" feature of its online community. This builds on the "social graph" that Facebook announced 3 years ago, which maps the explicit and implicit relationships among members based on their profiles, timelines, and behaviors within that community.

Facebook's offering has stoked the popular buzz for graph analysis, but it's far from the only appearance of this technology in the mainstream consciousness. The Internet is swarming with discussions of graphs in every possible big-data, data-science, digital-marketing, search-optimization, and other application context. Often, graphs are touted as some sort of secret sauce in new consumer-facing cloud services. In addition, the business world continues to adopt graph technology, encouraged by its proven value in anti-fraud, influence profiling, behavioral segmentation, customer experience management, and other applications.

However, I like to think there's an even simpler reason why graphs have become so popular: they're beautiful to behold. Pictures are worth a thousand words, and graphs, even when they're as densely packed as the Milky Way galaxy, generate some of the most gorgeous analytic visualizations you will ever see. Graph visualizations have an almost magnetic pull on your imagination. If it leverages solid data, a plausible model, and an interactive visualization tool, a graph can immerse you in an open-ended session of navigation, manipulation, and exploration.

Your aesthetics may be different from mine, but I find them mesmerizing. For example, here's a "citation graph" from IBM Research (

Here's the social network graph of one individual, the famed mathematician Paul Erdos, who had no permanent address and was a house guest of one friend after another (

And here's a social graph (

You can graph almost any complex data set in ways that are not only lovely to look at but also help the human mind pull out meaningful patterns in a way that pure data, numbers, or other visualizations rarely do. Interactive exploration is where graph analysis truly comes to life, but a static graph can be a thing of deep beauty and meaning in its own right. To see what I mean, type these keywords into Google, click "images" in the bar at the top of the page, and behold the diverse graph visualizations associated with each domain: relationship graph, influence graph, behavioral graph, experience graph, location graph, network graph, affinity graph, and semantic graph.

Even when a graph analysis application eschews pictorial "nodes and connections" visualization, it can still be quite compelling. For example, here's the grid/outline format that Facebook uses to render social graphs.

Graphs draw their power from concise visualization. Even when it balloons to include myriad nodes and connections, a well-crafted graph can focus your mind beautifully on meaningful patterns in the underlying data.