An Association Rule with Split Aggregate Merge Algorithm for Insurance Risk Management

Nowadays, Insurance risk management is crucial for minimizing losses in terms of economical perspective. Insurance risk is potential for adverse events that may occur during the policy term which impact the financial stability and ability in paying claims. Despite the rising of insurance buyers, many insurance companies still lack a detailed claim analysis system which leads to delayed settlements, inaccurate assessments and cooperating policyholder satisfaction. To overcome these issues, the methos of Association Rule with Split Aggregate Merge (ARSAM) is proposed to analyses the risk of insurance and incorrect assessment of policyholder’s risk profiles. Also, identify potential risks as well as make informed decisions to optimize their operations improve the individual risk profiles. The proposed ARSAM initially employes with the data collection from The Insurance Company (TIC) Benchmark dataset and extract relevant features from the claim data, which is implemented by the association rule; whereas SAM includes a risk assessment component to identify key metrics from the input data. The proposed ARSAM presents greater results than the existing methods of reciprocal Weibull extension for modeling extreme value distributions on insurance risk management in terms of accuracy (86.52%), precision (91.32%) and recall (90.21%) and f1 score (98.98%) respectively.