Data Mining in the Insurance Industry{0}


By Angel M.

As the world moves towards an era of “Internet of Things” companies are trying to find different ways to produce accurate and effective information in the insurance industry. In the years, past client’s information was poorly documented on paper and was kept in folders and filing cabinets. Today, we can see a major transition of the same information being documented online and stored in databases or data warehouse for easier access. Tech companies have produced software such as, data mining tools that allow insurance companies to sort through their information within seconds. The process will allow insurance companies to cut down on insurance fraud and increase operational efficiency.

The benefits of data mining in the insurance industry will not only help cut down on fraud claims, but also help identify new customers, increase sales, and improve human relation within companies. Data mining tools help identify policies that were issued on wrong or false information provided. Insurance companies can scan through this data information to see when peak hours of consumer purchasing to allocate themselves with more employees during those hours to eliminate excessive backlogs of transactions. Allowing companies to increase staff while improving human relations within their company. Additionally, with the increase in sales the insurance companies will also be able to take advantage of an increase in revenues while still providing their customers the best product they can.

With every opportunity, there comes challenges and data mining in the insurance industry is no different. Some of challenges faced in this industry include cost, time issues, and weakness of the data collected. Companies look to cut cost in any way possible, so investing in new software where revenue is not produced short-term makes them hesitant from the beginning. Data mining software can be extremely expensive and a long-term investment that needs constant attention with vital updates for the information to be processed correctly and be accurate as much as possible. Companies see it as a waste of money and time, because of the amount of resources needed to process the data such as, having the right software and employing properly skilled employees. A survey conducted by Gartner Group of 2,000 data warehouse projects found that less than 20 percent of the projects that were started were subsequently completed with a successful implementation; in the survey, 20 percent never got past the initial planning stage and another 60 percent did not complete the project after initiating the effort (Lampe, J. C., & Garcia, A.). Even after the data is collected insurance companies are still hesitant on the data collected, because of the data possibility being termed as “dirty-data.” “Dirty-data” means, data that it is incomplete or erroneous, which is data stored in a database that contains errors. This collection of “dirty data” leaves companies wondering if data mining was the right choice for their company from the start.

Tools and software at an insurance companies’ disposal come from top companies such as, International Business Machines (IBM), Statistical Analysis System (SAS), and Frontier Solver. These top companies have created data mining software that processes data at a faster rate that any one individual can process. Allowing insurance companies to identify, investigate, report, and prevent insurance fraud. Data mining software allows MetLife Insurance Company to identify policyholders who are committing rate evasion by lying about where they live or where they garage their cars in order to pay a lower premium (Lampe, J. C., & Garcia, A.). An upside to preventing insurance fraud is helping customers get better assistance and cutting costs at the same time. Maccabi Health Services, an Israeli HMO, uses data mining as part of its efforts towards continuous improvement; top management uses data mining to improve care quality while saving on costs (Lampe, J. C., & Garcia, A.).

In conclusion, data mining software will not entirely stop insurance fraud, but at least insurance companies will have the software needed to crack down the processing of fraudulent claims. Besides, cracking down fraud claims using data mining software insurance companies can also put their customers first by providing them with whatever consumer good or service they may have to offer.

References
Dorn, C. (2004). Data mining technology helps insurers detect health care fraud.National Underwriter.Life & Health, 108(39), 34-34,39.
Galfond, G. (1997). Data mining can unearth a competitive edge. National Underwriter, 101(40), 10-10,37.
Lampe, J. C., & Garcia, A. (2004). DATA MINING: AN IN-DEPTH LOOK. Internal Auditing, 19(2), 4-20.