<![CDATA[Public services in population administration play a fundamental role in ensuring citizens' access to legal identity documents and essential administrative services. The implementation of the Population Administration Information System (SIAK) has significantly improved service efficiency and data integration across administrative units. However, persistent challenges such as fluctuating service demand, uneven workload distribution, and limited data-driven evaluation mechanisms continue to affect overall service quality. Traditional perception-based evaluation methods, such as SERVQUAL, are considered subjective and insufficient for capturing operational dynamics in real-world service environments. This study aims to evaluate the quality of family card issuance services using a data-driven approach based on the FP-Growth algorithm. The methodology encompasses several stages, including data preprocessing, discretization using equal-frequency binning, transformation into binomial transaction format, frequent pattern mining, and association rule analysis employing support, confidence, and lift metrics. The dataset consists of daily service records categorized into four service types, namely YBS, 3in1, Death, and Urgent services. The results reveal that YBS and Death services form stable and dominant service patterns with consistent workload contributions, while 3in1 and Urgent services exhibit more dynamic and variable behavior requiring adaptive management strategies. The generated association rules yielded confidence values reaching up to 1.00 and lift values greater than 1, indicating strong and meaningful interdependencies between service categories. These findings offer practical insights for improving service responsiveness and operational performance. In conclusion, this study confirms that a data mining approach effectively supports objective service quality evaluation and evidence-based decision-making in optimizing resource allocation and operational planning.]]>
