Over the past few weeks I have been discussing the use of graph models for analyzing interconnectivity and how entity characteristics can be inferred in relation to links and connections. While we looked at the social network domain for identifying influential individuals within a social community, there are numerous other opportunities for examining the relationship between entities and identifying actionable knowledge. Here are three common examples:

• Mobile telecommunications network optimization: Mobile telecommunications companies seek to make the best use of their network in a number of ways, including ensuring the quality and continuity of calls (so that calls are not dropped in the middle of connections), providing equal distribution of bandwidth (to prevent any particular section of the network from being overloaded), and eliminating dead zones (where there is little or no connectivity).
• Cybersecurity monitoring: Cyber attacks are becoming increasingly more severe than the common distributed denial of service attack, which frequently becomes the cover for more dastardly attacks designed to extract critical business information and intellectual property or to raid financial accounts under the noses of the security professionals. Graph analytics models deployed on big data platforms not only are able to manage a real-time image of massive streaming NetFlow, DNS and IDS data, it enables continuous monitoring for connections and relationships indicative of ongoing or even imminent attacks. This allows security analysts to rapidly identify potential cyber-threats and notify key players who can take immediate actions to prevent a breach.
• Healthcare effectiveness analysis: The motivation to improve quality of healthcare includes the ability to absorb many different types of data – medical histories, clinical information, imaging results, laboratory test results, physician interactions, preferred prescriptions and patient accountability for taking those medications. Increasingly, providers are looking at using graph analytics for assessing many similar medical histories managed within a graph model that not only links patients to physicians, medications and presumed diagnoses, but also allow a provider. Providers can rapidly scan through the graph to discover therapies used with other patients with similar characteristics (such as age, clinical history, associated risk factors, etc.) that have the most positive outcomes.

In essence, any environment in which the links among a community of entities can form patterns that can lead to positive business impact is a candidate for graph analytics model. More examples include logistics (linking production facilities to warehouses to retail locations by truck, rail and air routes), public utilities (such as energy and water services), public safety (identifying areas that would benefit from an increased police presence to reduce crime), and homeland security (looking for terrorist cells). In a future series we will begin to examine the graph model in greater detail.