Fraud Claim Detection Framework

-

There are several fraud claims in insurance industry. The manufacturing industry also suffers with similar fraud claims. The manufacturers sells the product in the market with warranty or guarantee of good quality. They also promise to replace or repair the product according to the agreed policy. Based on the extent of risk and product life pricing of the selling product is decided.  Manufacturers receive several warranty/guarantee claims in a year. Many of them are fraud claims. Identifying the authenticity of claims is rigorous and laborious process. But in the era of the machine learning and data analysis the cost and time of above process can be reduced.  I am discussing here a process that can be used to identify the veracity of the claim. This process framework is based on the machine learning algorithms. I believe that this framework has capability to reduce the time and cost required to investigate any claim manually. 

The process flow of the claim review as follows:
On-line interface: Customer can claim using on-line interface which will save tremendous amount of time spent in paper work by both customer and OEM manufacturer. Customer can check status of the claim on the real time basis.

Data Storage: Customer data is stored for future use. The data stored can be used to enhance the fraud detection statistical model. It is also linked with the customer warehouse data. It means it contents the product information, not necessarily customer information.

Data Processor: The data processor will check the appropriateness of the information provided by the customer for the claim. In case of any data anomalies, customer will be notified on the real time basis through on-line interface. This service can be further enhanced by using SMS/email service.

Real Time Processor:  When information provided by customers satisfies the pre-decided level of data coherency, the information supplied is processed by the Real Time Processor. This is a system which comprises of statistical model for fraud detection, text mining techniques, Claim Score model and business rule. Statistical Models are built based on the historical data of claim. Fraud detection model of the third party can also be used like tools of Advance Analytical Consulting Group. This framework can also be developed in house by implementing machine learning algorithms.



1. Valid Claim: - Customer will be informed through on-line interface and all the details will be passed for processing of the claim.

2. Suspicious Claim: - Customer information will be passed for manual review

3. Fraud Claim: -   The claim will be rejected and customer will be informed the decision of rejection with sufficient reason(s).

Suspicious claim will be again classified into two categories Valid and Fraud Claim after manually reviewing the information.  Customer will be informed accordingly.

Data Storage will have feedback mechanism to keep record of status of each claim. The data stored can be used to enhance the statistical model on periodical basis.

Business benefit
1. The customer on-line interface will make the process completely paperless and user friendly.
2. The reduction in fraud claims.

3. Reduction in cycle period of processing the claim because of reduced manual review and paperless work.

Comments

Post a Comment

Popular posts from this blog

Solution for ERROR: Some character data was lost during transcoding in the dataset

-
When I search for the mentioned error I get the link solution provided administration oriented but developers do not have access.     http://support.sas.com/kb/52/716.html Please see my solution below proc options option=config; run; proc options group=languagecontrol; run; /* Show the encoding value for the problematic data set */ %let dsn=item_information_may16; %let dsid=%sysfunc(open(&dsn,i)); %put &dsn ENCODING is: %sysfunc(attrc(&dsid,encoding)); /*Renaming item desc file  (encoding=any) allowed reading****************************/ data tmp.item_info_curr; set hc.item_information_may16 (encoding=any); run; /*gave error*/ data item_info_curr; set item_information_may16; run;
Read more

How to check whether a SAS dataset exist or not and throw an error in the log ?

-
/* if the data set exists, then conditionally execute the step(s). %sysfunc checks the availability of the data in the library location. It means it checks historical sas data. It generates the system error if the not is data available at the defined library  */ %let dsname=&dataval.MKT_DATA;  /**** declaring data set name with library*****/ %macro warning1(name); %if %sysfunc(exist(&name)) %then %do; Proc printto log="&logfile.\WarningLog..txt" new;     /*** printing log in the text file at specifile location &logfile****/ run; %Put WARNING :- &name does not exists; Proc printto  ; Run; %end; %else %do; Proc printto log="&logfile.\WarningLog.txt" ; run; %put &name exists; Proc printto  ; Run; %end; %mend ; %warning1(&dsname); /**** Pros:  If any sas process is dependent on the existence of the data then we can use this macro***/ /**** Cons:  *Even if the SAS data set is available with zero observatio
Read more

2018 plan for getting expertise in Machine Learning and Deep Learning

-
Machine Learning and deep learning are next frontier in the world of innovation. These skill sets are high in demand and demand is going to increase further as we are moving towards a world with connected systems. Although I am experienced professional with SAS having multiple SAS certifications and  using SAS for programming, predictive model building, optimization and data visualization for more than seven years, I will be foolish I do not recognize how important it is to adopt open source platforms for innovation.  2018 plan to acquire expertise in these are are 1. Choose a programming language and I have chose python as my language of preference  2. Do hands on practice to be efficient in Pandas and Numpy. These two libraries are useful and very important for data exploration,wrangling and data cleaning. I find Kernals in Kaggle and below blogs and books to be helpful tutorial for  Pandas  a. https://www.dataquest.io/blog/pandas-tutorial-python-1/  b.https://www.d
Read more