Abstract For insurance firms, client churn, or the absence of customers, is a major problem because it affects revenue and sales. This study suggests a unique method for recognizing client churn in an insurance firm called modified monarch butterfly optimized random forest (M2BO-RF) to tackle this issue. This approach aims to improve the precision and efficacy of customer churn prediction using both RF and modified monarch butterfly optimization (M2BO). The foraging behavior of monarch butterflies, renowned for their impressive navigational abilities and an effective quest for supplies, served as the model for the MBO algorithm. The fundamental components of the MBO technique are incorporated into the proposed MMBORF algorithm, which changes the conventional RF technique. We performed trials on a dataset from an insurance company to assess the efficacy of the MMBORF method. The experimental findings showed that, in terms of prediction precision, Recall, accuracy, and F1-score, MMBORF surpasses other algorithms. Insurance businesses may create focused retention strategies and deploy resources effectively due to the algorithm's ability to detect probable churners successfully. Ultimately, this can improve customer satisfaction, lower customer churn, and contribute to insurance businesses' success.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|