Claim Missing Document
Check
Articles

Found 22 Documents
Search

Optimizing Insurance Customer Segmentation with C4.5 Decision Tree Algorithm Sigit Candra Setya; Moch. Iswan Perangin-angin; Marsono Marsono; Asyahri Hadi Nasyuha; Lucia Nugraheni Harnaningrum
Journal of Information System Research (JOSH) Vol 6 No 4 (2025): July 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v6i4.7358

Abstract

Insurance companies rely on premium payments as their primary source of revenue. However, economic instability often causes delays in premium payments, impacting revenue recording. This study applies the C4.5 Decision Tree algorithm to classify insurance customers based on premium amount, age, income, and claim history, thereby improving product recommendations. The research utilizes data mining techniques to analyze customer attributes and generate decision rules for optimal insurance product selection. The findings indicate that customers with a premium of IDR 500,000 are best suited for PRUMed Cover (PMC), while those with IDR 1,000,000 are recommended PRUCritical Benefit 88 (PCB88). For customers with IDR 750,000, additional factors such as age and income level influence the recommended insurance type. The entropy and information gain calculations identify premium amount as the most significant attribute for decision-making, followed by age, income, and claim history. By implementing this method, insurance companies can enhance customer segmentation, streamline product selection, and optimize marketing strategies. The transparent and interpretable decision tree structure ensures regulatory compliance while improving customer satisfaction. Future research should explore additional variables, such as behavioral data and regional trends, and compare C4.5 with other classification algorithms like Random Forest or Support Vector Machines (SVM) to enhance accuracy and scalability.
Evaluation of Green Marketing Strategy Using Fuzzy AHP-Based Decision Support System Marsono Marsono; Asyahri Hadi Nasyuha; Evi Rosalina Widyayanti; Meng-Yun Hadi Chung
Sinkron : jurnal dan penelitian teknik informatika Vol. 10 No. 3 (2026): Article Research July 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v10i3.16277

Abstract

Growing environmental awareness among consumers has encouraged companies to integrate ecological considerations into their marketing activities. However, many firms still find it difficult to determine which green marketing strategy should be prioritised, because the decision involves multiple conflicting criteria and a high degree of subjective human judgment. This study designs and applies a decision support system based on the fuzzy analytic hierarchy process (FAHP) to evaluate and rank green marketing strategy alternatives for a consumer goods company. The decision problem was structured into a three-level hierarchy consisting of the main goal, five evaluation criteria, namely green product, green price, green promotion, green distribution, and green corporate image, and four strategy alternatives. Expert judgments were gathered through pairwise comparison questionnaires using linguistic variables that were converted into triangular fuzzy numbers. Chang’s extent analysis was applied to compute the fuzzy synthetic extent and the degree of possibility, and the priority weights were normalised and verified through a consistency check (CR = 0.095, below the 0.10 threshold). To strengthen the validity of the recommendation, the alternative ranking obtained from FAHP was cross-validated against the Simple Additive Weighting (SAW) and TOPSIS methods. The results indicate that green product is the most important criterion (0.327), followed by green promotion (0.277) and green corporate image (0.210), while sustainable packaging is identified as the most preferred strategy alternative (0.294). All three methods produced an identical ranking, and a sensitivity analysis confirmed that the ranking remained stable under reasonable variations in the criteria weights. The proposed decision support system, whose architecture and interface are also presented, offers a transparent, consistent, and reproducible tool that helps managers allocate resources to the most effective green marketing strategy.