Segmentation of Pax Ecclesia Junior High School Students Using Deep Embedded Clustering Algorithm to Optimize Marketing Strategy

EasyChair Preprint 15920

Abstract

The competition among schools in attracting prospective students is becoming increasingly intense, necessitating a data-driven marketing strategy. This study applies the Deep Embedded Clustering (DEC) algorithm to group Pax Ecclesia Junior High School students based on specific characteristics, such as residence location and school origin. The evaluation was conducted using the Silhouette Index to assess clustering quality. The results identified three main groups: (1) students living around the school, (2) students residing at medium to long distances, and (3) students from specific elementary schools with close ties to Pax Ecclesia Junior High School. Based on this segmentation, more effective marketing strategies can be implemented, such as direct promotion for students near the school and digital marketing for prospective students from distant areas. The implementation of this data-driven strategy enhances the school's competitiveness and promotional effectiveness. Furthermore, this study demonstrates that the DEC algorithm can serve as a reliable tool for student segmentation.

Keyphrases: Analisis data, Data Mining, Sistem Informasi, Strategi promosi digital, machine learning

@booklet{EasyChair:15920,
  author    = {Herianata Windosari Barasa and Habibullah Akbar and Gerry Firmansyah and Agung Mulyono},
  title     = {Segmentation of Pax Ecclesia Junior High School Students Using Deep Embedded Clustering Algorithm to Optimize Marketing Strategy},
  howpublished = {EasyChair Preprint 15920},
  year      = {EasyChair, 2025}}