{"id":4248,"date":"2016-11-01T08:03:58","date_gmt":"2016-11-01T08:03:58","guid":{"rendered":"http:\/\/www.odbms.org\/blog\/?p=4248"},"modified":"2016-11-01T08:03:58","modified_gmt":"2016-11-01T08:03:58","slug":"on-fraud-detection-medicaid-and-the-insurance-industry-interview-with-charles-kaminski-jr","status":"publish","type":"post","link":"https:\/\/www.odbms.org\/blog\/2016\/11\/on-fraud-detection-medicaid-and-the-insurance-industry-interview-with-charles-kaminski-jr\/","title":{"rendered":"On fraud detection, Medicaid, and the insurance industry. Interview with Charles Kaminski Jr."},"content":{"rendered":"

“From my perspective,\u00a0data quality\u00a0is paramount to an evolving market. When the quality of data improves in a market, both insurance carriers and consumers can make better decisions. “–Charles Kaminski\u00a0Jr.<\/strong><\/p><\/blockquote>\n

I have\u00a0interviewed Charles Kaminski<\/strong> Jr.<\/strong>, Sr. Architect at LexisNexis Risk Solutions<\/a>. Main topics of the interview are the\u00a0technological challenges the insurance industry is currently facing,\u00a0fraud detection, and how\u00a0to effectively use \u00a0predictive analytics.<\/p>\n

RVZ<\/p>\n

Q1. What is your role at LexisNexis Risk Solutions?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>I am a Sr. Architect at LexisNexis Risk Solutions. I\u2019ve worked for LexisNexis Risk Solutions for about 7 years. My primary\u00a0responsibility is international expansion for the Insurance vertical. I also work on enterprise initiatives, new technologies, new product development, patents & intellectual property, and acquisitions. From time to time I work with RELX<\/a> sister companies when they need help. The RELX Group is our parent company.<\/p>\n

Q2. How is the life insurance industry evolving?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>My view is somewhat specific to the international markets I serve. From my perspective, data quality<\/a> is paramount to an evolving market. When the quality of data improves in a market, both insurance carriers and consumers can make better decisions. As that happens, the vast majority of consumers and other players in that market benefit. This isn\u2019t limited to the life insurance industry, but I see it happening there as well.<\/p>\n

Q3. What are in your opinion, the main technological challenges the insurance industry is currently facing?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>Each market around the global tends to have its own nuances that don\u2019t apply to any other market. An entity in one market (such as a bank, an aggregator, or a software house) may play a different role or no role at all in another market. Regulations, government involvement, and industry support also vary greatly. I see this in auto, life, and health verticals. These factors create different challenges from one market to the next. But, there are a few themes that seem to exist regardless of market.<\/p>\n

Insurance<\/a> carriers around the globe tend to utilize a healthy mixture of old and new technologies. The technology leaders in this industry are generally more risk adverse when compared to other, less regulated, industries. Also, workflows on the carrier side can be very complex. The primary technological challenge to new product development is understanding customer and vendor technology roadmaps and the implied assumptions in those roadmaps. Understanding the entities in a market as well as their roadmaps is key to being successful.<\/p>\n

Q4. Cross-industry fraud is defined by a fraud case where the perpetrator\u2019s activity touches multiple industries and organizations, habitually exploiting system gaps. Is using data and analytics the solution to fraud detection?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>A product person might better answer if using data and analytics is \u201cthe\u201d solution to fraud detection<\/a>. I can tell you it is a very effective solution. Big data<\/a> can cross boundaries and tell unique stories like no other tool. Companies that reign supreme in crossing those boundaries are the ones that have the technical capabilities to analyze big data with ease and the creative people to ask questions no one else is thinking to ask. One interesting story I can relay here is from work others at LexisNexis have done. It comes from someone I\u2019ve shared a stage with a number of times, so I\u2019m very familiar with the story.<\/p>\n

LexisNexis Risk Solutions was asked to help a US state agency identify potential Medicaid<\/a> fraud. Medicaid fraud<\/a> is big business with lots of money changing hands. For any state agency with limited resources, it\u2019s never a question of finding enough fraud to prosecute. It\u2019s always a question of finding the big fish to fry.<\/p>\n

This US state agency in question could only share the addresses of people using Medicaid and nothing more.
\nJust a list of addresses is not much to go on. But with the right tools, it\u2019s a good start: Why is someone at one address registering a number of really expensive cars? Why is someone at another address registering a rather expensive boat?
\nWhy is someone at yet another address who owns a Medicaid processing business and buying multiple multi-million-dollar condos, why are they possibly on Medicaid?<\/p>\n

Some of these will no doubt be coincidence and I\u2019m oversimplifying this by not mentioning some additional and rather complex analysis. I\u2019m sure you get the idea though. Ultimately you have an interesting list of addresses scored and ordered in terms of where you might want to take a closer look. But that\u2019s not where this story ends. That scored and ordered list is just where this story starts to get interesting.<\/p>\n

With a big-data system geared towards analytics, we can take that list and overlay relationship data on top of it.
\nYou can build relationship data from all kinds of sources — who\u2019s married or ever been married to whom, previous neighbors who lived near each other, jointly-registered assets, various public records from business dealings, etc.
\nWhen we overlay who knows who, multiple circles start to form. People who don\u2019t know each other are in these circles and at the center of many of these circles (connecting them together) are people who weren\u2019t in the original address list.
\nThose folks in the center of those circles are the big fish to take a closer look at. Many of these people in the center are the generals recruiting lieutenants to commit the fraud for them. These generals do this so they can stay below the radar.<\/p>\n

That\u2019s the interesting part of this story. It\u2019s a story of how big data and analytics can take you from just a list of addresses to some big fish in the center of a fraud ring.<\/p>\n

Q5. Drew Whitmore, Senior Director, Insurance Global Alliances, LexisNexis\u00ae Risk Solutions, said: \u201cInsurance carriers need innovative core policy and claims management solutions integrated with industry-leading data and analytics to meet their business objectives and deliver on promises of exceptional customer experience,\u201d Why do you believe that a single point of entry to these data and analytic solutions is the best option for insurers\u2019 technology resources and workflow processes?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>Insurance workflows<\/a> can be very complex. Products that support these workflows can have complex interface.\u00a0To a technologist, success with a single-point-of-entry strategy is very clear.
\nSuccess is when we release a new product but 90% of the single-point-of-entry-interface doesn\u2019t change.
\nFurther still, success is when the technologist on the other side, the employee of the customer, knows exactly what is going on with the new product. Success is when a technologist on the other end of the interface says, \u201cI get what LexisNexis is doing with this.\u201d That technologist also benefits when he or she needs to discuss the new product with a legal departments or internal auditors because those groups will already be familiar with the interface.<\/p>\n

Q6. What is the LexisNexis Risk Solutions telematics data and analytics platform? And how is it used in the Insurance industry?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>The telematics platform<\/a> is a horizontally scalable, high performance, big data and analytics platform. It and the associated data is used by carriers who want to understand driving behavior as well as a number of other attributes associated with a policy. Because the platform is format agnostic, carriers have quite a bit of flexibility to use our solutions or bring their own to the table.<\/p>\n

I was part of the original team bringing telematics solutions to market. We considered a number of different problems to solve, prototypes, and solutions in those early days. We went through a number of iterations before settling on our first telematics solution. That initial product enabled telematics for carriers by using a consumer\u2019s smart phone, an ODB2 dongle<\/a>, and LN\u2019s scalable data analytics systems<\/a> to store and analyze the data. A dedicated telematics team continues to expand our telematics offerings. I\u2019m no longer involved day-to-day.<\/p>\n

Q7. According to a Gartner report* referencing its 2015 CIO Study, \u201ceighty-seven percent of CIOs agree that there is a shift to predictive analytics from reporting in their organizations, and 79% believe that the greatest value and insight will come from active experimentation informed by data rather than the passive analysis of data.\u201d What is your take on this?<\/strong><\/p>\n

Charles Kaminski<\/strong>\u00a0Jr.:\u00a0<\/strong>Big data and predictive analytics<\/a> are powerful tools that have transformed a number of industries. For insurance, they are a must. But these tools are now being adopted by a number of other industries and they are sometimes misapplied. There are a number of cautionary case studies in business news where these capabilities were brought into an organization with high cost and high expectations but the investment provided negative returns. Wikibon is reporting that most enterprises expect a return of $3.50 per dollar spent on big data systems but that the actual return to date is more like $0.55 per dollar spent.<\/p>\n

My take on this is twofold. First, if you are looking to bring big data and predictive analytics in house, then spend some time choosing the right first business case with a low cost and a low bar to success. This gives you greater flexibility to find scarce resources around big data and predictive modeling, prove out your technology, and fine tune your assumptions. Also, be sure the resources you engage with have experience getting positive returns using big data and analytics.
\nSecond, if you are an executive looking to drive improvements with these tools and you do not currently have a predictive analytics engine, then consider broader trends first. Twenty years ago businesses goals were being managed through results.
\nSince then there has been a shift towards driving business and organizational improvements using lead measures and lead indicators. This doesn\u2019t necessarily mean predictive analytics. These lead measures and lead indicators can be developed and iterated over quickly without big-data and complex analytics. They can then be used to drive improvements across an enterprise. This can be done before tools such as big data and predictive modeling are introduced.
\nThere are people and firm that can help businesses get started immediately with comparatively low costs.<\/p>\n

————————
\nCharles Kaminski<\/strong> is a Sr. Architect for LexisNexis Risk Solutions. Charles was part of the team that open-sourced the LexisNexis big data platform, HPCC Systems, which is the backbone of LexisNexis Risk Solutions. He now focuses on global markets and international expansion for the company\u2019s Insurance business. Charles has worked for NASA in their Solar System Exploration Division, Accenture’s Financial Services vertical, and was an entrepreneur before joining LexisNexis Risk Solutions. Charles lives outside of Atlanta with his wife and children.<\/em><\/p>\n

————————<\/p>\n

Resources<\/strong><\/p>\n

*Gartner, \u2018Market Trends: Targeting Global Life and P&C Insurers in 2015,\u2019 23 April 2015, Derry N. Finkeldey<\/a><\/p>\n

– LexisNexis Risk Solutions Elevates Insurance Customer Experience with New Active Risk Management Solution3\/1\/2016<\/span><\/a><\/p>\n

–\u00a0LexisNexis Risk Solutions Expands Relationship with Duck Creek Technologies<\/a><\/p>\n

–\u00a0Big Data Revolution: What farmers, doctors and insurance agents teach us about discovering big data patterns.\u00a0Authors:<\/em>\u00a0Rob Thomas, Patrick McSharry<\/a><\/p>\n

–\u00a0Introduction to HPCC (High-Performance Computing Cluster).\u00a0Authors: Anthony M. Middleton, Ph.D. LexisNexis Risk\u00a0Solutions and Arjuna Chala, Sr. Director Operations, LexisNexis Risk Solutions.ODBMS.org,\u00a0FEBRUARY 19, 2016<\/a><\/p>\n

—\u00a02016 HPCC Systems Engineering Summit \u2013 Community Day<\/a><\/p>\n

Related Posts<\/strong><\/p>\n

–\u00a0MarkLogic Case Study: Hannover Re<\/a><\/p>\n

–\u00a0Ethical Risk Assessment of Automated Decision Making Systems,\u00a0By\u00a0Steven Finlay,\u00a0Head of Analytics at HML<\/em>. ODBMS.org\u00a0FEBRUARY 23, 2015<\/a><\/p>\n

Follow us on Twitter: @odbmsorg<\/a><\/strong><\/p>\n

##<\/p>\n","protected":false},"excerpt":{"rendered":"

“From my perspective,\u00a0data quality\u00a0is paramount to an evolving market. When the quality of data improves in a market, both insurance carriers and consumers can make better decisions. “–Charles Kaminski\u00a0Jr. I have\u00a0interviewed Charles Kaminski Jr., Sr. Architect at LexisNexis Risk Solutions. Main topics of the interview are the\u00a0technological challenges the insurance industry is currently facing,\u00a0fraud detection, […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[66,1024,1029,1009,1030,796,210,1027,1026,1025,1031,1032,700,1033,1028],"_links":{"self":[{"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/posts\/4248"}],"collection":[{"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/comments?post=4248"}],"version-history":[{"count":7,"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/posts\/4248\/revisions"}],"predecessor-version":[{"id":4271,"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/posts\/4248\/revisions\/4271"}],"wp:attachment":[{"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/media?parent=4248"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/categories?post=4248"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.odbms.org\/blog\/wp-json\/wp\/v2\/tags?post=4248"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}