Ayasdi and The Power of Topological Big Data Analysis
When it comes to finding out who the winning companies will be for any disruptive technology, it’s always a good idea to look at where the smart money is going. The only way you’re going to accurately track this information real time for privately held companies is by using the powerful data platform from CB Insights which is the most comprehensive database of startups there is. CB Insights recently published a list of the 10 highest funded startups developing core artificial intelligence technologies. The first startup on this list was Sentient Technologies which we highlighted in a previous article. The second most funded startup today is a company called Ayasdi.
Founded in 2008, Silicon Valley startup Ayasdi has taken in $106 million in funding from big-name investors that include Khosla Ventures, Draper Fisher Jurvetson, General Electric, Kleiner Perkins Caufield & Byers, and Citigroup. Ayasdi claims to have developed the most powerful technique known to man for advanced analytics of big data and complex data. The technology is called “topological data analysis” and here is how it works.
If you are trying to analyze a group of databases, you would probably look to build a “data warehouse” which is just a connected set of databases that you can now query as if it were one database. This approach requires that you identify common keys between each database in order to link them all together. This is what you might call a bottom-up approach and it takes a great deal of time and analysis work. Ayasdi takes a much simpler approach by simply identifying each database as a node and then connecting them all together in a giant network of connections. Ayasdi then uses topology, a sub-field of mathematics that looks at shapes, to recognize patterns in the data that are relevant. This top-down approach makes it much easier to find complex relationships that you may have never found through a traditional bottom-up approach.
Ayasdi is targeting two niche applications for their “topological big data analysis” technology; healthcare and financial services. Here’s one application in healthcare called “claims denial reduction”.
Let’s say you’re a health care provider and the majority of your patients use insurance so you need to fill out a claim for every patient. According to Ayasdi, “rejected claims represent hundreds of billions of dollars in lost revenue for healthcare organizations each year, costing them about 5% of their net revenue streams“. With the average cost of resubmitting a claim at $251, 65% of all rejected claims are just written off. With Ayasdi’s topological data analysis tools, you can now begin to identify which claims you should resubmit for $251 a pop which are most likely to go through the second time around. Trying to figure out those patterns manually could take an expensive consulting team weeks to discover, and then may just rectify only a small part of the problem. For all you math aficionados who might say that this sounds like clustering, read this blog post from Ayasdi that says why it’s not.
In addition to claims denials reduction, Ayasdi is also targeting something called “clinical variation management”. What this means is that Ayasdi will quickly analyze all the electronic medical records (EMRs) and financial data for a healthcare organization representing millions and millions of records, to objectively compare the costs and impact of all variations in care delivery. In other words, you can quickly see how you might do things more efficiently to save costs or optimize profits.
In addition to healthcare, Ayasdi offers solutions for the financial sector that include stress testing, market returns and risk forecasting, anti-money laundering, fraud detection, and risk modeling. Ayasdi makes the bold claim that their machine intelligence is as much as 1000x more effective than previous systems and processes used for advanced analytics as their clients are finding their modeling process times reduced from thousands of hours to minutes. These extraordinary claims are backed up by an impressive client list that includes Lockheed Martin, The U.S. Federal Reserve, Citi, Merck, Johnson & Johnson, and many others. Ayasdi even used topological data analysis to provide a glimpse into what may be powering the Donald Trump engine, the visual depiction of which is as follows:
Given Ayasdi’s all-star list of investors, their strong and varied client base, and their prominence as the second most well funded artificial intelligence startup out there, this Company is one to watch in this space.
If you enjoyed this article, then sign up for our free newsletter - Nanalyze Weekly. About every week, we'll send you a simple summary of all our new articles. If you didn't enjoy this article, share it on Twitter and tell everyone how much you hated it.