Smart data discovery company Bottlenose Inc has announced the addition of two major new sources to its ever-expanding data library: LexisNexis, which stores offline data from government, academic, and industry sources; and Flashpoint, which stores data from the deep web and the dark web.

Bottlenose co-founder and CEO Nova Spivack told me that these new integrations represent a huge boost in power for the platform, allowing businesses to better anticipate new threats and opportunities as they arise.

Spivack explained that by adding the data from LexisNexis and Flashpoint to Bottlenose’s existing sources, the platform can better detect new patterns as they form by comparing a variety of signals that might seem insignificant on their own.

“If you’re only looking at social, or the Dark Web, or newswires, there may be a blip, but it’s unlikely to be noticed,” Spivack told me. “But by cross-correlating and overlaying all the weak blips across all the data, we can see stronger signals begin to emerge in aggregate. Weak signal detection is a major need right now, detecting anomalies before they’re obvious, and being able to monitor them as they grow.”

According to Spivack, Bottlenose has been able to help businesses anticipate customer support issues, prepare for potential cyber threats, forecast opening weekends for films, and more.

The Three Vs of Big Data

The Three Vs of Big Data is a model commonly used to describe the challenges faced by the data industry. The Vs include Volume, which is the raw amount of data available, Variety, which is the range of data sources, and Velocity, which is the speed at which data is produced and analyzed.

According to Spivack, Volume has been the primary challenge for the Big Data industry for some time, but he believes that is no longer the case.

“Data is growing exponentially, and the Volume issue has been the driver over the past decade of big data technology, pouring vast amounts of data into the lake,” Spivack said. “But today, nearly 80 percent of data is unstructured or semi-structured, and Velocity and Variety of data are the new problems, and they are not well solved yet.”

“Companies need to be able to separate the signal from that vast amount of noise, especially as you feed in more and more data. Bottlenose can separate out what’s important. We elevate the data from raw data, to analytics, to intelligence. And we don’t gather then compute, but instead we do continuous analytics.”

He added, “Systems like Hadoop and even Spark that are batch processed are just too slow for this era of stream data. The analytics and discoveries are live and real time.”

photo credit: NATS Press Office via photopin cc