Description: Up until recent past, the artificial intelligence portion of data science was looked upon cautiously due to its history of booms and flops . In the latest stream of events, major improvements have taken place in this field and now deep learning, the new leading front for Artificial Intelligence, presents a promising prospect for overcoming problems of big data.

Description: This article goes through some basic data exploration techniques to show how different approaches look.

Description: Presented at May 1985 Spring Meeting. This session discussed the differences in the reporting of product line and source of earnings results between insurers and banks and the practical problems which exist in interpreting gain by source or product-level.

Description: Panel discussion about ways can insurance carriers and health care coalitions work together, what data are coalitions requesting from health carriers, how health carriers respond to those requests, and implications for health carriers.

Description: This paper is a proof of a conjecture that appeared in a letter to the Editor of The Actuary. From the Actuarial Research Clearing House 1991 Vol.

Description: It has been twenty years since DeWit 1982 first applied fuzzy logic FL to insurance. That article sought to quantify the fuzziness in underwriting. Since then, the universe of discourse has expanded considerably and now the FL methodologies include used fuzzy set theory, fuzzy numbers and fuzzy arithmetic, fuzzy inference systems, fuzzy clustering algorithms, fuzzy linear programming, fuzzy regression, and hybrids involving other soft computing technologies. This article presents an overview of these studies.

Description: A new health insurance pricing model takes external market prices into account, as well as regulatory rate restrictions and other traditional pricing factors. Rather than being assumed at the outset, lapse rates emerge from the model, along with assessment spiral & loss ratio projections. Under the model optimal pricing strategy depends on whether the class being rated has a healthier than average mix of risks.

Description: Although rate structures where low-cost cohorts subsidize high cost cohorts may be justified from a actuarial, marketing, or legal standpoint, the subsidization creates risks which must be understood and monitored. This article presents the concept of subsidization signature to measure and interpret the effects of rate subsidization, and to assess the sustainability of rating structures and the sensitivity of business mix assumptions.

Description: This session explores the far-reaching changes to how clinical data will be sourced, processed and consumed in many payer use cases including risk adjustment, quality measurement, care management, claims adjudication, prior authorizations, and other payer functions. Robust, interoperable data exchange is a prerequisite for value-based care. Clinical data will be increasingly accessible to augment administrative (claims) data via standardized Fast Health Interoperability Resources (FHIR, pronounced 'fire'), at a fraction of the cost of a accessing a traditional medical record and in real-time. On May 1, 2020, HHS issued a final rule implementing interoperability and the patient access provisions of the 21st Century Cures Act of 2016, although enforcement has been delayed into 2021 because of the COVID-19 pandemic. The new regulations promote, “HHS’s vision and strategy for achieving a health information ecosystem within which all individuals, their personal representatives, their health care providers, and their payers are able to send, receive, find, and use electronic health information in a manner that is appropriate, secure, timely, and reliable to support the health and wellness of individuals through informed, shared decision-making, as well as to support consumer choice of payers and providers.”Once enforcement begins, CMS and the Office of the National Coordinator for Health Information Technology (ONC), have the authority to mandate adoption of interoperability requirements, and also penalize payers and providers for non-compliance or the delay and withholding of data.This ‘digital transformation’ of the clinical data ecosystem has far-reaching implications for enabling technologies including Natural Language Processing (NLP) and entire segments of services and vendors. It will eliminate many departmental barriers, duplication and error sources for health plans and impact every department and the underlying economics of delivering and managing care.

Description: Can we improve upon traditional methods of risk measurement and risk selection? Panelists discuss potential applications of predictive modeling, including individual underwriting, group rating and provider reimbursement. They focus on ways in which predictive modeling can be used to improve risk selection, experience rating and provider reimbursement. A session at the Health/Pension Spring Meeting held in New Orleans, Louisiana June 15-17, 2005. From the Record of the Society of Actuaries, Vol. 31, No. 2.