Understanding postal delivery areas in the Republic of Korea using multiple unsupervised learning approaches

Abstract

Changes in household composition and the residential environment have had a considerable impact on the features of postal delivery regions in recent years, resulting in a large increase in the overall workload of domestic postal delivery services. In this paper, we provide complex analysis results for postal delivery areas using various unsupervised learning approaches. First, we extract highly influential features using several feature-engineering methods. Then, using quantitative and qualitative cluster analyses, we find the distinctive traits and semantics of postal delivery zones. Unsupervised learning approaches are useful for successfully grouping postal service zones, according to our findings. Furthermore, by comparing a postal delivery region to other areas in the same group, workload balancing was achieved.

Improved Deep Learning-Based Energy Management Strategy for Battery-Supercapacitor Hybrid Electric Vehicle With Adaptive Velocity Prediction

“…Feature engineering involves the transformation of data into features that state the problem of the predictive model well to improve model performance and reduce complexity [38]. The driving cycle dataset has only one feature of velocity, v, this hampers the prediction performance of the LSTM model as many input features can improve its accuracy.…”

Special issue on smart cities and its applications

“…Moreover, changes in household composition and residential environment greatly affect the characteristics of postal delivery areas, increasing the overall workload of domestic postal delivery services. The fifth paper in this special issue, [10] "Understanding Postal Delivery Areas in the Republic of Korea Using Multiple Unsupervised Learning Approaches" by Han and others, investigated the overall workload of postal delivery services using various unsupervised learning approaches in the Republic of Korea. In the study, the authors applied multiple clustering methods, as well as feature engineering, to obtain optimal grouping of postal delivery areas and workload balancing of postal delivery services.…”

Minería de procesos en la Ingeniería de Software

“…Otro aspecto interesante se muestra en [13] donde se hace foco en los cambios en la composición de los hogares y el entorno residencial de los últimos años en la República de Corea que ocasionaron un gran impacto en las regiones urbanas, lo que produjo un aumento considerable de la carga de trabajo de los servicios de entrega postal. En el documento se realizan diferentes tipos de análisis complejos para áreas de entrega postal utilizando aprendizaje no supervisado.…”

“…Se eligió Anaconda Navigator 12 una interfaz gráfica que permite administrar extensiones y lanzar paquetes de Anaconda y entornos de desarrollo. Puntualmente se utilizó el IDE Spider 13 para hacer el desarrollo por ser uno de los más utilizados en Python para hacer desarrollos e investigación científica. La figura 5.5 ilustra el desarrollo de la prueba de concepto en el IDE Spider.…”