Click here to go to the previous page
Tightening the Noose: Innovations in Claim Fraud Detection Webinar
Program Code:
190
PRESENTER
(S):
Dr. Karthik Balakrishnan has more than 15 years of experience in building learning systems and predictive models for challenging problems in both industry and academia. He has published more than 20 papers on neural computation, evolutionary algorithms, and neuro-cognitive modeling, and is the coauthor of an MIT Press book on evolutionary synthesis of intelligent agents. He is a strong proponent of the "toolkit" approach — leveraging multiple modeling and analysis methodologies to craft pragmatic solutions. In his current role, Dr. Balakrishnan oversees all predictive analytics initiatives at IIA, including property/casualty and mortgage fraud analytics.
Dr. Balakrishnan has more than nine years of predictive modeling experience in property/casualty insurance, building and directing the development of solutions for marketing (customer segmentation, channel/producer segmentation, opportunity scoring), underwriting/pricing (personal auto, homeowners), and operations (claim fraud detection, subrogation identification, premium audit prediction). He has also managed data warehousing and direct mail/marketing functions at several companies. He began his career at Allstate Research and Planning Center and led analytics organizations at Obongo (a subsidiary of AOL) and Fireman's Fund Insurance Company. He has been with IIA since June 2007.
|
Thomas Mulvey has been in the business of insurance investigations for more than 27 years. In his current position, he serves ISO as the national director of SIU and claim services. He is also the coordinator of ISO’s Insurance Fraud Management Conference. Before joining ISO, Mr. Mulvey was the national SIU director for the Prudential Property & Casualty Insurance Company. He also had responsibilities in the area of corporate investigations as a director at Prudential in their worldwide operations.
Mr. Mulvey has authored numerous state antifraud plans, SIU operational manuals, industry “best practice” studies, and SIU-related articles for trade periodicals. He has also served as a police and fire academy trainer and has been a frequent presenter at international, national, and regional antifraud conferences and seminars.
Mr. Mulvey is a graduate of Providence College and holds a master’s degree in criminal justice administration. He has also held the designations of Certified Insurance Fraud Investigator and Certified Fraud Examiner.
Mr. Mulvey is an active member of the New Jersey Special Investigator’s Association. He has served on committees and working groups with the Coalition Against Insurance Fraud, the National Insurance Crime Bureau, the C-COR SIU Committee, and the New Jersey Office of the Insurance Fraud Prosecutor.
|
Description
b>Program Description
This session will provide a detailed introduction to the R statistical computing package. R is both an open source language and computing environment with very strong statistical and graphical capabilities. The R project benefits from the active participation of a rapidly expanding global network of users and academics, and is quickly gaining traction within the global actuarial community.
Starting with installation of the software and continuing with an introduction to the R programming language, the session will provide a strong foundation from which attendees will be able to explore R for themselves. The session will first introduce such basic mechanics of the R environment as loading and manipulating data, reading and writing script files, and data exploration. The session will also cover some of the more sophisticated capabilities of R, such as building and graphically analyzing generalized linear models. One or more property-casualty case studies involving real data will be included.
Note: It is not necessary to install or use R prior to or during the webinar.
Intended Audience
This webinar will be most suitable for those who have a keen interest in the R programming language but who have little or no experience with it. It is intended to be introductory in nature and will therefore contain some of the minutiae of using a command-line interface.
Intended Audience
CAS Members and Non-Members are welcome to attend. This session may be of particular interest to those actuaries who would like to explore predictive analytics as a tool to assist in detecting fraudulent claims. Additionally this session will be of interest to claims and special investigative units.